Chapter 2 - Flagellar Motility in Bacteria: Structure and...

C H A P T E R T W O

In

IS

*

ternati

SN 1

Divis

Flagellar Motility in Bacteria:

Structure and Function of

Flagellar Motor

Hiroyuki Terashima,* Seiji Kojima,* and Michio Homma*

Contents

1. In

onal

937

ion

troduction

Review of Cell and Molecular Biology, Volume 270 # 2008

-6448, DOI: 10.1016/S1937-6448(08)01402-0 All rig

of Biological Science, Graduate School of Science, Nagoya University, Nagoya, Japa

Else

hts

n

40

1
.1. M olecular architecture of flagella 40
1
.2. G ene regulation 44
1
.3. F lagellar assembly 46
1
.4. R egulation of rotation 48
2. B
asal Structure of Flagella as Motor 50
2
.1. B asal body 50
2
.2. E xport apparatus 52
2
.3. S witch complex 53
2
.4. M otor complex 58
3. T
orque Generation 61
3
.1. In teraction between stator and rotor 61
3
.2. Io n-binding site 64
3
.3. Io n specificity 66
3
.4. A ssembly of functional motor 67
4. M
olecular Physiology of Motor 68
4
.1. T orque–speed relationship 68
4
.2. S teps in rotation of motor 70
4
.3. F luorescent imaging of motor 72
5. C
onclusion 74
Refe
rences 74
Abstract

Bacterial flagella are filamentous organelles that drive cell locomotion. They

thrust cells in liquids (swimming) or on surfaces (swarming) so that cells can

move toward favorable environments. At the base of each flagellum, a revers-

ible rotary motor, which is powered by the proton- or the sodium-motive force,

vier Inc.

reserved.

39

40 Hiroyuki Terashima et al.

is embedded in the cell envelope. The motor consists of two parts: the rotating

part, or rotor, that is connected to the hook and the filament, and the nonrotat-

ing part, or stator, that conducts coupling ion and is responsible for energy

conversion. Intensive genetic and biochemical studies of the flagellum have

been conducted in Salmonella typhimurium and Escherichia coli, and more than

50 gene products are known to be involved in flagellar assembly and function.

The energy-coupling mechanism, however, is still not known. In this chapter, we

survey our current knowledge of the flagellar system, based mostly on studies

from Salmonella, E. coli, and marine species Vibrio alginolyticus, supplemented

with distinct aspects of other bacterial species revealed by recent studies.

Key Words: Proton motive force, Sodium motive force, Energy transduction,

Vibrio alginolyticus,Salmonella typhimurium, Escherichia coli. �2008Elsevier Inc.

1. Introduction

1.1. Molecular architecture of flagella

The flagellum consists of three parts: the filament (helical propeller), thehook (universal joint), and the basal structure (rotary motor) (Fig. 2.1). Thelargest part of the flagellum is the filament, a helical structure whose shapecan vary among different helical forms, a phenomenon termed polymor-phism (Asakura, 1970). This polymorphic alteration of flagellar shape isassociated with phase variation (Iino, 1969). When the cell swims, theflagellar filament serves as a screw propeller to convert rotary motion ofthe motor into thrust (Berg and Anderson, 1973). In Salmonella, it grows toa length of around 15 mm and is composed of as many as 30,000 copies of asingle protein named flagellin (Minamino and Namba, 2004). Some bacte-ria, for example Vibrio, have several closely related flagellins that form thefilament (McCarter, 2001). The flagellin subunits (FliC in Escherichia coli andSalmonella) are self-assembled to form a hollow concentric double-tubularstructure (inner and outer tubes) consisting of 11 protofilaments, which arearranged approximately parallel to the filament axis (Mimori et al., 1995;Morgan et al., 1995). Formation of a helical structure is achieved by amixture of the protofilaments of two distinct conformations, the R- andL-type, distinguished by their helical handedness right or left (Asakura,1970; Calladine, 1978). Each protofilament switches between these twoconformations by responding to a variety of factors including pH, ionicstrength, mechanical stress, and mutations (Kamiya and Asakura, 1976;Macnab and Ornston, 1977). Later, X-ray fiber diffraction studies revealedslightly different subunit packing between the R- and L-type, whose repeatdistances are 51.9 and 52.7 A, respectively (Yamashita et al., 1998). To

Smooth swimming

Smooth swimming

Tumbling

A

B

Cap

Filament

Joint

Hook

L ringP ring

MS ringSwitch complex

Exportmachinary

BasalbodyRod

OMPG

IM

Mot complex

C

Figure 2.1 (A)Behaviorof bacterial cells. (B)Electronmicrographof flagella isolatedfrom Salmonella typhimurium. (C) Schematic diagram of flagellar structure of Gram-negative bacteria.

Structure and Function of Flagellar Motor 41

understand the polymorphic switching mechanism, the crystal structure offragment containing most of the flagellin protein (F41 fragment of FliC) wassolved at 2 A (Samatey et al., 2001). In the crystal, F41 fragments of theR-type conformation form protofilaments arranged in an antiparallel


manner, and simulation using atomic model revealed that a small, distinctconformational changes in the b-hairpin in D1, the domain that contributesto outer tube structure, is responsible for the conformational switchingbetween L- and R-type protofilament. The crystal structure of F41 frag-ment lacks the D0 domain that forms inner tube structure. Later, electroncryomicroscopy and helical image analysis made it possible to build com-plete atomic model of the R-type filament structure including D0 domainof flagellin, revealing that intersubunit hydrophobic interactions in theinner tube (domain D0) make the filament structure mechanically stable,and the diameter of central channel is only 2 nm (Yonekura et al., 2003).This central channel serves as a transport pathway of flagellins that willpolymerize at the tip of the growing filament (Fig. 2.2).

The base of the filament is connected to the short tubular structure calledthe hook, which is thought to function as a universal joint to smoothlytransmit the torque produced by the motor to the filament. The hookstructure of Salmonella is composed of about 120 copies of a single proteinFlgE and its length is controlled at 55 (�6) nm (Hirano et al., 1994). Thecore structure of the hook protein was solved by X-ray crystallography, andits atomic model was docked onto the density map of the hook obtained byelectron cryomicroscopy (Samatey et al., 2004).

The junction between hook and filament consists of the two proteins,FlgK (HAP1) and FlgL (HAP3) (Homma and Iino, 1985; Ikeda et al., 1985).About 13 molecules of each protein are present in each flagellum (Ikedaet al., 1987). Mutational studies suggested that the junction acts as a buffer-ing structure connecting two filamentous structures (hook and filament)with distinct mechanical properties (Fahrner et al., 1994).

The proximal end of the hook is connected to the basal body structure,consisting of the rod and three coaxially mounted rings, termed as MS, P,and L ring. The MS ring is embedded in the cytoplasmic membrane andmade of a single protein FliF (Ueno et al., 1992), the P and L rings areassociated with the peptidoglycan layer and the outer membrane, respec-tively, and are composed of FlgI and FlgH (Homma et al., 1987; Schoenhalsand Macnab, 1996). The rod structure is composed of three proximal rodproteins FlgB, FlgC, FlgF, and a distal rod protein FlgG, and fully traversesthe periplasmic space. The L and P rings together form a quite rigidassembly resistant to stringent chemical treatments, and the LP-ring com-plex is believed to act as a molecular bushing for the flagellar axial structure(Akiba et al., 1991). The basal body of Gram-positive bacteria is composedof only the MS ring and rod, and the LP ring is not present (Kobayashi et al.,2003), probably because Gram-positive bacteria do not have the outermembrane but have a thick peptidoglycan layer. When the basal body isisolated with more gentle treatment, a drum-shaped structure, called C ring,was found on the MS ring facing the cytoplasm (Francis et al., 1994). It iscomposed mostly of FliM and FliN proteins (Thomas et al., 2006; Zhao

A

B

C V283

A191A298

A392D140

N132E153

N56R450

K160

D3 D2

D2a

D2b

D1K177

N406N100

N315

T346

Figure 2.2 (A) Polymorphic structure of the flagellar filament. The filaments ofL-type straight, normal, curly, and R-type straight are shown by the left to right in thisorder. (B) The flagellar filament structure revealed by electron cryomicroscopy.The end-on view of the cross section (left) and the side view of long segment (right).(C) The backbone trace of the flagellin (FliC) molecule of Salmonella typhimurium.The figureswere kindly supplied byK.Namba.


et al., 1996a,b). These proteins, together with FliG which is located beneaththe MS ring, have been known to form a complex, referred to as the switchcomplex. Mutations in each of these three proteins cause defects in switch-ing the rotational direction of the motor (Yamaguchi et al., 1986a). Theyare also important for rotation, and mutational studies revealed that FliGmost closely participates in torque generation (Lloyd et al., 1996). Carefulpreparation of the basal structure further revealed a central protrusionwithin the C ring, that is probably the export apparatus essential forassembly of flagellum (Katayama et al., 1996).


1.2. Gene regulation

More than 50 genes are required for flagellar formation and function(Macnab, 2003). Because the flagellum is such a big organelle, a largeamount of energy is consumed during the assembly process. Bacteria dealwith this problem by developing highly organized and regulated systems forflagellar assembly. Its characteristic feature is that the flagellar gene regula-tion temporally and tightly couples to the assembly process. Here we surveythis regulatory mechanism in Salmonella, which has been most extensivelystudied. For further details of flagellar gene regulations in Salmonella entericaserovar typhimurium and E. coli as well as the other bacteria, such as Caulo-bacter cresentus and Vibrio spp., see the reviews cited in Chilcott and Hughes(2000), McCarter (2004), Wolfe and Visick (2008), and Wu and Newton(1997).

In Salmonella, the flagellar/motility/chemotaxis genes constitute a reg-ulon, and they are arranged in hierarchical order into three classes, early,middle, and late (Fig. 2.3) (Kutsukake et al., 1990). At the top of thehierarchy is a single operon (class 1 master operon) containing the flhDCgenes (Liu and Matsumura, 1994). The FlhC and FlhD proteins form aheterotetrametic complex FlhC2FlhD2 that direct s70-RNA polymerasecomplex to activate transcription from class 2 promoters upstream of the

Class1

Stimuli

FlhDC

FliT

2

1

2

2

2

2

23

23

23

3

3

3

3

3

3

FliZ

flhDC

flgBCDEFGHIJ

fliFGHIJK

fliLMNOPQR

fliEmotAB cheAW

tar cheRBYZ

hin fljBA

fliB

fliCfliDST

fliAZYFliA(σ28)

FlgM(anti-σ28)

flgA flgMN

flhBAE

flgKL

σ70

σ70

Class2 Class3

Figure 2.3 Regulation of transcription of the flagellar regulon.The flagellar operonsare indicated by arrows. The numbers on the arrows show the class of transcriptionalhierarchy in the flagellar regulon. Descriptions of the transcriptional regulation andthe function of the gene products are described in text.


middle gene operons. The flhDC operon is tightly regulated under thecontrol of a number of global regulatory signals such as cAMP-CRP, heatshock, DNA supercoiling, growth phase, surface-liquid transition, ClpXPprotease, and c-di-GMP (reviewed in Chilcott and Hughes, 2000; Wolfeand Visick, 2008). Expression of flhDC is also linked to the cell cycle,showing that flagellation and cell cycle are interdependent processes (Prussand Matsumura, 1996, 1997).

More than 30 middle gene products from class 2 operons are primarilyrequired for the structure and assembly of the hook and basal body, andinclude regulatory proteins that control transcription of the genes from class3 operons. Class 3 operons encode for proteins required late in the assemblyprocess, including flagellin, hook-associated proteins (HAPs), stator com-ponents, and chemosensory systems. Expression of class 3 operons is posi-tively controlled by FliA and negatively by FlgM (Gillen and Hughes, 1991;Kutsukake and Iino, 1994). FliA is the flagellum-specific transcription factors28 (Ohnishi et al., 1990), and it leads s28-RNA polymerase complex totranscribe class 3 late gene operons, whereas FlgM is an anti-s28 factorspecific for FliA (Ohnishi et al., 1992). Since filament formation requireslarge resources of the cells, the stage to initiate expression of class 3 operonsis a critical checkpoint for flagellar gene regulation. Actually, what happensin vivo is that the late genes are not transcribed until the assembly of hook-basal body structures has been completed. Mutations in any one of thehook-basal body genes prevent the transcription of class 3 operons to avoidunnecessary late gene expression that would cause an energy drain in thecell. This remarkable coupling between structure and gene expression isachieved by the balance of FliA and FlgM. FlgM inhibits FliA for theexpression of class 3 operons until hook-basal body completion, at whichtime FlgM is secreted out from the cells through hook-basal body structureand released FliA turns on transcription of class 3 operons to achievefilament formation and motor assembly (Hughes et al., 1993; Kutsukake,1994). Such a controlled expression is coupled to the ordered secretion ofeach gene products by the flagellum-specific export apparatus, putativelylocated inside of the MS ring, to complete the self-assembly of flagella.

As described above, sensing hook-basal body completion followed byFlgM secretion is such a critical checkpoint; Salmonella utilizes anotherfactor involved in negative regulation in FlgM secretion. This gene, flk(Karlinsey et al., 1997), also called as rflH (Kutsukake, 1997), was identifiedas the factor that allows expression of class 3 operons only when mutated instrains defective in LP ring assembly. Flk is a cytoplasmic-facing proteinanchored to the inner membrane by a single, C-terminal transmembranedomain, and it was revealed that turnover of FlgM was increased in flkbackground due to FlgM secretion into periplasm where it is degraded,suggesting that Flk prevents premature secretion of the FlgM into peri-plasm, thus acting as a braking systems for the flagellar export system


(Aldridge et al., 2006). Loss of only Flk does not show any phenotyperelated to motility, and flk gene is not located in the flagellar regulon, sothe role of Flk in the wild-type cell is still unclear (Karlinsey et al., 1997;Kutsukake, 1997).

1.3. Flagellar assembly

As described in Section 1.2, flagellar assembly is tightly coupled to the geneexpression, and the monitoring system at the stage of hook completionallows cells to achieve efficient and economical filament formation. Thesequence of events in flagellar assembly has been extensively studied inSalmonella and E. coli by studying partial structures of flagella from mutantsdefective in a certain flagellar gene (Kubori et al., 1992; Suzuki and Komeda,1981; Suzuki et al., 1978). In general, assembly starts at the inner structure ofthe basal body then proceeds to the outer ones (Fig. 2.4). The first builtstructure is the MS ring and proximal rod, which is formed by a singleprotein FliF (Ueno et al., 1992). The MS ring is the core structure of therotor and is embedded in the cytoplasmic membrane. The C ring attaches on

FlgE

FlgD FlgHFlgl FlgBCFG

FliF

FliGMN

FlhAB,FliOPQR

FlgD

HAP1HAP3 HAP2

Flagellin

HAP2(FliD) Cap

Filament

Joint

Hook

Basal body

Switch complex

Exportmachinary

HAP3(FlgL)HAP1(FlgK)

FlgE

L ring(FlgH)P ring(Flgl)Rod(FlgBCFG)MS ring(FliF)C ring(FliGMN)

FlhAB,FliOPQR

Flagellin(FliC)

Figure 2.4 Morphological assembly of bacterial flagellum.The model of pathway iscreated based on the assumption that the structure is built from a simple to complicatedform.The flagellar morphogenesis is thought to begin with the formation of MS ringstructure followed by the assembly of proteins for axial structures, each of which issecreted by the flagellum-specific export apparatus (except FlgA, FlgI, and FlgH).


the cytoplasmic face of MS ring (Francis et al., 1994). The C ring containsmostly two switch proteins FliM and FliN (Zhao et al., 1996a,b), and isassociated with MS ring via another switch/motor protein FliG, whichprobably contributes to a part of the face of MS ring (Thomas et al., 2006).Assembly of these three proteins on the basal body requires the MS-ringplatform, and mutations give rise to the nonflagellate phenotype. Inside theMS ring, there is the flagellum-specific export apparatus, visualized in freeze-fracture images as a protrusion inside the C ring (Katayama et al., 1996).When the export apparatus is established in the flagellar base is still unclear.Details of the export apparatus are described in Section 2.2.

After the export apparatus is constructed, structural proteins for the basalbody, expressed from class 2 operons, are secreted through the exportapparatus in the order described below. First, the proximal rod, composedof FlgB, FlgC, and FlgF, is added on the MS ring, probably in this order(Homma et al., 1990). FliE is needed for this assembly, joining FliF, andproximal rod as an adaptor (Minamino et al., 2000b). Then the distal rod,made of FlgG, is assembled on the proximal rod. Formation of the rodrequires FlgJ, which is exported to the periplasmic space via export appara-tus and acts as a cap on the growing rod to facilitate the polymerization atthe tip (Kubori et al., 1992). FlgJ also has a muramidase activity at itsC-terminal half, hydrolyzing the peptidoglycan adjacent to the MS ring toallow the rod to penetrate the peptidoglycan layer (Nambu et al., 1999).Then the P ring (made of FlgI) is formed around the distal rod, followed byL-ring (FlgH) formation (Chevance et al., 2007). FlgI and FlgH are notsecreted through the export apparatus, but through the Sec pathway using asignal sequence at their N-termini (Homma et al., 1987). P-ring formationrequires the Dsb system, which is involved in intramolecular disulfide bondformation in the periplasm (Dailey and Berg, 1993). FlgI protein containstwo cysteine residues important for protein stability (Hizukuri et al., 2006).P-ring formation also requires the FlgA protein that acts as a periplasmicchaperone, assisting a polymerization reaction of FlgI into the P ringthrough FlgI–FlgI interaction (Nambu and Kutsukake, 2000). The hookassembles next, from about 120 copies of FlgE proteins at the distal end ofthe growing hook, with the aid of hook-capping protein FlgD. Hookelongation proceeds to the well-controlled length of 55 � 6 nm by asophisticated export switching mechanism (Hirano et al., 1994). Afterhook reached to the defined length, FlgD dissociates from the tip of thehook, then replaced by the three HAPs, FlgK, FlgL, and FliD in this order(Homma and Iino, 1985). Addition of FlgK and FlgL is facilitated by thechaperone FlgN (Fraser et al., 1999), whereas that of FliD is facilitated byanother chaperone FliT (Bennett et al., 2001). Finally, the FliC filamentsubunits (flagellin) are inserted at the distal end (Iino, 1969). FliD acts as acap to facilitate the filament elongation by inserting each FliC subunitbetween the FlgL and FliD zones, with rotary cap mechanism revealed by


electron cryomicroscopic observation (Yonekura et al., 2000). Mutants thatlack FliD cannot polymerize the filament and excrete flagellin monomersinto the culture (Homma et al., 1984). However, when purified FliD isadded to this mutant, flagellins stop leaking out and start polymerizing onthe hook (Homma et al., 1986). Using this system, growth rate of thefilament was observed and revealed that initial growth rate is about30 nm/min, which corresponds to one flagellin incorporated per second,suggesting that to reach the 10 mm long of the filament of wild-type cells, ittakes several generations (Ikeda et al., 1993).

Vibrio alginolyticus has a single polar flagellum, so components requiredfor polar flagellum are localized to a single cell pole. The mechanism fordirecting the MS ring (or to initiate MS-ring assembly) at the pole hasremained unknown, but recent studies revealed that two proteins, FlhF andFlhG, are somehow involved in the process (Kusumoto et al., 2006). Almostall of the cells of an flhF null strain do not have a polar flagellum, whereas anflhG strain has multiple flagella at a pole. Overproduction of FlhF in thewild-type strain increased the number of polar flagella, whereas excess FlhGreduced it, indicating that these two proteins function in opposing ways.Although the flhFG double null strain also showed almost no flagella, a verysmall but significant fraction of the cells possesses several flagella at the lateralposition (Kusumoto et al., 2008). These results suggest that FlhF functions inpolar location of the flagellum. This idea is supported by a study in Pseudo-monas putida, possessing polar flagella, which showed that an flhF mutantexhibits a peritrichously flagellated phenotype (Pandza et al., 2000).

FlhF and FlhG show similarity to FtsY and MinD, respectively. FtsY is acomponent of the prokaryotic SRP receptor (Luirink and Sinning, 2004)and MinD is a cell division inhibitor (Shapiro et al., 2002). MinD shows astructural similarity to Ffh (Cordell and Lowe, 2001), which is a prokaryoticSRP that forms a complex with FtsY to function in a signal recognitiontargeting pathway for protein secretion at the membrane (Focia et al., 2004).Therefore, FlhF may function together with FlhG in the same manner withFtsY/Ffh system to locate flagellar components, possibly the MS-ringprotein FliF, to the cell pole. The direct interaction between FlhF andFlhG has been suggested (Kusumoto et al., 2008). A recently reported crystalstructure of FlhF from Bacillus subtilis revealed a dimer formation of FlhFand significant structural similarity of FlhF to FtsY/Ffh, supporting the ideadescribed above (Bange et al., 2007). However, many features of the polarlocalization mechanism remain unknown.

1.4. Regulation of rotation

Most flagellar motors are reversible rotary machines, able to rotate bothclockwise (CW) and counterclockwise (CCW) (Silverman and Simon,1974). Rotational switching completes very quickly, within only 1 ms


(Kudo et al., 1990). Rotational direction is controlled by environmentalstimuli, such as pH, temperature, and chemicals like sugars and amino acids.Methyl-accepting chemotaxis proteins (MCPs) sense these stimuli andtransmit signals to the motor through a two-component phosphorelaysignaling cascade (Fig. 2.5). When a repellent signal is sensed by theMCP, autophosphorylation activity of the CheA protein, associated withMCP on the cytoplasmic side, is activated and a histidine residue of CheA isphosphorylated. Then this phospho group is transferred to the Asp residueof the response regulator CheY. Phosphorylated CheY protein (CheY-P)then associated with the motor to trigger CW rotation. On the other hand,when an attractant signal is sensed by the MCP, autophosphorylationactivity of CheA is repressed, so that the level of CheY-P decreases andthe motor rotates in its default direction, CCW. The molecular mechanismsof MCP function and two-component signaling are reviewed elsewhere(Armitage, 1999; Parkinson et al., 2005).

The target of CheY-P in the motor is the switch complex, composed ofFliG, FliM, and FliN. As described above, FliG/FliM/FliN complex is alsocalled ‘‘the switch complex’’ because mutations in these proteins causedefects in switching the CCW/CW rotation in response to tactic stimuli.FliM functions most directly in regulation of the switching frequency bybinding to CheY-P (Welch et al., 1993). This binding of CheY-P to FliMprobably changes the FliG–FliM interaction, and causes movement of theC-terminal domain of FliG (FliGC) that interacts with the stator proteinMotA, thereby altering the rotor–stator interface to switch the direction ofrotary motion.

Signal

Receptor

Flagellar motor

ATP

ADP

CheW

CH3

CheA

CheA-p

CheY-p

CheYCheZ

FliGFliMFliNp

ppi

Figure 2.5 Schematic diagram of signal transduction of Escherichia coli chemotaxis.The chemoreceptors are embedded in the cytoplasmic membrane and localized at apole. Chemotaxis substances or ligands bind to the receptor and the signals are transmit-ted into a cell and are transferred through the two-component phosphorelay systemviathe Che proteins.The phosphorylated CheYcan bind to FliM and the direction of themotor rotation is changed.


Studies of intracellular level of CheY-P in a single cell that causes switch-ing fromCCW to CW revealed that switching is a highly cooperative event,showing a Hill coefficient of about 10, suggesting that chemotactic signal isamplified within the switch (Cluzel et al., 2000). Fluorescence resonanceenergy transfer-based observation of CheY interaction with FliM by usingCFP-FliM andCheY-YFP showed that binding of CheY-P to FliM is muchless cooperative thanmotor switching (Hill coefficient of 1.7 � 0.3) (Sourjikand Berg, 2002). This result was further supported by in vitro biochemicalstudies showing that CheY-P binding to the isolated intact switch complexwas not cooperative (Hill coefficient is around 1) (Sagi et al., 2003). Theseresults indicate that the chemotactic signal is amplified within the switch, butsubsequent to the CheY-P binding to FliM.

Some bacteria respond to tactic stimuli using modes, other than direc-tional switching. Rhodobacter sphaeroides has a unidirectional flagellar motorthat alternates between CW rotation and brief stops, during which thebacterium is reoriented by Brownian motion and changes in flagellar fila-ment morphology (Armitage and Macnab, 1987). In the case of Sinorhizo-bium meliloti, the motor also rotates unidirectionally in the CW direction andswimming cells respond to tactic stimuli by modulating the flagellar rotaryspeed (Schmitt, 2002). The marine bacterium V. alginolyticus has dual flagel-lar systems, Naþ-driven polar flagellum (Pof ) and Hþ-driven lateral flagella(Laf ), and their switching modes are different: the Laf motor rotates unidi-rectionally in CCW and responds to tactic signals by slowing down, whereasthe Pof motor turns in both directions (Kojima et al., 2007). In each casedescribed, tactic signals are transmitted through the two-component signal-ing pathway, and CheY-P association to the motor modulates rotation.

2. Basal Structure of Flagella as Motor

2.1. Basal body

Structural features and the components of the basal body are described inSection 1.1. The supramolecular complex of the basal body is constructedby 7 kinds of proteins and �130 molecules in total are assembled (Macnab,2003). Aizawa et al. (1985) established a method to isolate the basal body ofSalmonellawith high yield and purity. Using this method, but greatly modifiedto allowC-ring isolation, the structure of the flagellar basal body andC ring ofSalmonella had been investigated by the single particle analysis using negativelystained- or cryo-electron microscopy (Francis et al., 1994). A resolution wasobtained at 20 A that is used for building the three-dimensional reconstitution


of images (Suzuki et al., 2004; Thomas et al., 2001, 2006). The flagellar basalbody has a rotational symmetry with its axis in the center of the rod. Thecenter of the rod is a hollow tube and the flagellar components required foraxial structure are exported through this tube. The detailed single particleanalysis of the MS ring revealed an interface between the MS ring and rod(Suzuki et al., 2004). TheMS ring looks like a cylinder mounted on two disks.Suzuki et al. (2004) reported that the MS-ring structure can be divided intofive domains (C, P, S,M, andR), and discussed functions of each domain. Thenear-axial C and P domains are involved in the protein export and thought tochange its conformation by the associationwith the export apparatus and opena channel through the protein. The C domain associates with the mostproximal side of the rod and thereby the rod begins its assembly on the Cdomain. Although the S and M domains form the two disks characteristic ofthe MS ring, their functions are not clear. The R domain that forms cylinder-like part attaches to the outer face of the rod. It was speculated that theinteraction between the rod and R domain is important for the rod–MSring junction to release the twisting stress and symmetry mismatch. Thomaset al. (2006) has greatly improved resolution of images, by classifying particlesaccording to size and applying the averaging procedures appropriate for eachsymmetry class. This improvement allows us to see for the first time thedetailed feature of the C ring, not just a dumbbell-like structure. Theirimages revealed that the symmetry of individual M rings varies from 24-foldto 26-folds, whereas that of the C rings varies from 32-fold to 36-fold, with noapparent correlations between the symmetries of the two rings. The resolutionof these improved EM images is now good enough to allow the crystalstructures solved for the C-ring components to be docked into the map.The LP ring is an extraordinarily rigid structure and can maintain the ringmorphology even under stringent condition such as 7.5 M urea (Akiba et al.,1991). Single particle analysis revealed that the P ring looks like a dumbbell andseems to contact with a part of the rod, whereas the L ring seems not to contactthe rod (Stallmeyer et al., 1989).

MotX and MotY, which are required for the rotation of the polarflagellar motor of V. alginolyticus, are associated with the basal body andform an additional ring structure beneath the LP ring, termed T ring(Terashima et al., 2006). Partial T-ring structures were observed in theDmotX strain but not in the DmotY and DmotXDmotY strains, suggestingthat MotX associates with the basal body via MotY to form a completeT-ring structure. Stoichiometry of the MotX and MotY proteins in the Tring has not been determined yet. Further detailed observations of the Tring to reconstruct the three-dimensional images will be informative tounderstand their function.


2.2. Export apparatus

Most of the flagellar substructures are constructed beyond the cytoplasmicmembrane. Therefore, protein components, synthesized in the cytoplasm,must be exported across the inner and outer membranes to be assembled atthe appropriate final destinations. This is achieved by the flagellum-specificexport apparatus that resides inside the MS-ring structure (Macnab, 2004).This system has a character in common with the needle complex that worksfor the secretion of virulence factors by pathogenic bacteria (Cornelis,2006). Morphology of these two secretion machineries is quite similar toeach other, and they are now classified in the type III export pathway(Hueck, 1998). The flagellum-specific export apparatus exports proteinsubstrate in order without signal peptide cleavage. Most of the studies forthe export apparatus have been carried out in Salmonella, and here wedescribe its general overview.

The core of the export apparatus is composed of six transmembraneproteins, FlhA, FlhB, FliO, FliP, FliQ, and FliR, that form an exportchannel complex inside the MS-ring structure. Three cytosolic proteinsFliH, FliI, and FliJ are required for flagellum-specific export and interactwith the channel, thereby contributing a part of the export apparatus.Recent study revealed that flagellar C-ring protein FliN is also involvedin the export, so it is a part of the apparatus (Brown et al., 2005; Gonzalez-Pedrajo et al., 2006; McMurry et al., 2006; Paul et al., 2006). FlhA and FlhBhave a large cytoplasmic domain at their C-termini, where soluble compo-nents interact with. The export apparatus (or a part of it) has been visualizedas a protrusion inside the C ring, by the freeze-fracture image (Katayamaet al., 1996). However, only FliP and FliR have been detected in the basalbody preparations so far (Fan et al., 1997). FliI is an ATPase and its sequenceshows similarity to the b subunit of the FOF1-ATPase, a rotary motor thatdrives chemical reaction of ATP synthesis (Fan and Macnab, 1996).Recently, crystal structure of FliI is solved at 2.4 A, and revealed that itssimilarity is not only in the sequence (29% identity) but also in the structurallevel (Imada et al., 2007). FliI is a member of the Walker-type ATPasefamily, and it is thought to form a ring-shaped hexamer for protein export(Claret et al., 2003; Minamino et al., 2006). These lines of evidence lead toan attractive model that FliI hexamer functions as a motor, unfolding andthreading export substrates thought its central channel by cooperativeconformational changes of subunits, just as speculated for AAA ATPasecomplexes. ATPase activity of FliI is negatively regulated by binding toFliH, in a complex of FliH2FliI1 stoichiometry (Minamino and Macnab,2000a). FliH also binds to the hydrophobic patch of FliN, and theirinteraction would mediate efficient localization of FliI near the C-ringcomplex. FliJ is a general chaperone, preventing aggregation of exportsubstrates presumably by interacting with them (Fraser et al., 2003;


Minamino et al., 2000a). Interactions among export components have beeninvestigated by far-western method (affinity blotting), and it was found thatcytosolic components interact with each other and probably forming FliH/FliI/FliJ complex in some point, and the cytoplasmic domain of membranecomponents FlhA and FlhB interacts with all soluble components(Minamino andMacnab, 2000b). Affinity blotting experiments also indicatethat cytoplasmic domain of FlhA and FlhB interact with substrates, suggest-ing that these two proteins are involved in protein translocation. From theseresults, the outline of export is emerging: protein substrate is captured byFliH/FliI/FliJ complex without denaturation and transferred to the exportapparatus. In the same time, FliI docks to the cytoplasmic face of theapparatus, forming hexameric ring structure that has an enhanced ATPaseactivity released from FliH inhibition. Then the substrates are exportedthrough a central channel of apparatus, probably formed mainly by trans-membrane segments (TMs) of FlhA and FlhB. After crossing the innermembrane, substrates are assembled at the appropriate position in orderedfashion, from the innermost structure to the external ones.

How is this ordered export achieved? As discussed in Section 1.2,coupling of flagellar gene expression to the stages of flagellar assemblymakes it possible for the ordered export. But there is one moremechanism operating in the flagellum-specific export apparatus: thesubstrate-specificity switching (Kutsukake et al., 1994). The apparatus hastwo export substrate-specificity states, the rod/hook type and the filamenttype (Minamino and Macnab, 1999). Therefore, proteins forming the rodand hook structures are exported prior to the proteins required for filamentformation. When the hook structure reaches a certain length (ca 55 nm),FliK and FlhB sense this state, and the substrate specificity of the apparatusswitches from the rod/hook type to the filament type, causing the export ofa new substrate class (Ferris and Minamino, 2006; Moriya et al., 2006;Shibata et al., 2007; Waters et al., 2007). In addition, this switching leadsto FlgM export, followed by expression of the class 3 operons (Hughes et al.,1993; Kutsukake, 1994). It is reported that this switching is irreversible(Minamino et al., 1999).

2.3. Switch complex

The switch complex functions in the rotation/switching/assembly of theflagellum and is composed of the three kinds of proteins, FliG, FliM, andFliN (Fig. 2.6). Null mutants of each protein exhibit Fla� phenotype (non-flagellate), and point mutations give rise to Fla�, Che� (defective in chemo-taxis), andMot� (defective inmotility) phenotypes (Yamaguchi et al., 1986a,b). Stoichiometry of the FliG, FliM, and FliN in the C ring have beenreported to be 26, 34, and more than 100 copies, respectively (Francis et al.,1992; Suzuki et al., 2004; Thomas et al., 1999). Three-dimensional

Salmonella cell

H+-type

L ring (FlgH)

P ring (Flgl)

RodMotB

MotA PomA

PomB

T ring (MotX, Y)

HookNa+-type

FliG FliGC ring

Export apparatusMS-ring (FliF)

ca. 50 nm

FliM, N FliM, N

Vibrio cell

Figure 2.6 Cell body and flagellar basal structure of the proton- and sodium-driventype. The components emphasized by darker drawing are essential components fortorque generation. The stator of flagellar motor consists of MotA and MotB of Hþ-driven motor in Escherichia coli and of PomA and PomB of Naþ-driven motor inVibrioalginolyticus. For the Naþ-driven motor, additional components of MotX and MotY,which form aT ring in the basal body, are also essential. Ion flow through MotAB orPomAB complex is believed to be coupled with torque generation by the interactionbetween the stator component of MotA or PomA and the rotor component of FliG.Electron micrograph of osmotically shocked Salmonella cell to visualize flagellar baseis shown in left photograph. The flagellum of V. alginolyticus cell is observed andvisualized by the electronmicroscopy in right photograph.


reconstructions from electron cryomicrographs of the rotor revealed that theC ring displays �34-fold symmetry and the MS ring shows about 25-foldsymmetry (Thomas et al., 2006), and confirmed that most of the C-ringstructure was composed of FliM and FliN. It was known that the FliF bindsto FliG, and N-terminal 46 residues of FliG are required for the binding(Oosawa et al., 1994). Recent studies revealed more detailed FliG–FliMinteractions, and it will be discussed later (Brown et al., 2007). In addition,spontaneous mutants of FliF–FliG fusion were found to be functional(Francis et al., 1992). Therefore, the FliF:FliG ratio should be 1:1, and it is


consistent with FliG stoichiometry in C ring and the symmetry ofMS ring asdescribed earlier. By direct comparison of the ring structure made of FliF andFliF–FliG fusion protein obtained by electron cryomicroscopy and single-particle image revealed that a part of the FliG occupies the outer rim and faceof theM ring, whereas the remaining part is likely a part of the C ring (Suzukiet al., 2004). FliM interacts with FliG and FliN but not with FliF, and FliNdoes not interact with FliF (Oosawa et al., 1994). Therefore, FliM and FliNare probably located at the central and bottom position of the C ring,respectively.

Structural studies of the switch proteins have been undertaken by usingthe thermophilic bacteria Thermotoga maritima as a protein source, and crystalstructures of the functional domains have been determined (Fig. 2.7). ForFliG, middle and C-terminal domains of the protein (residues 104–335,termed FliGMC) correspond to two-third of the full-length protein (Brownet al., 2002; Lloyd et al., 1999). The middle domain (residues 115–190,termed FliGM) has a conserved surface patch formed by the residuesEHPQ125–128 and R160 (the EHPQR motif ), which is important forbinding to FliM (Brown et al., 2007). The FliGC (residues 198–335) has aconserved surface hydrophobic patch, which is also important for thebinding to FliM (Brown et al., 2007). A region near the C-terminus hasfive well-conserved charged residues that interact with those in the statorprotein MotA, and electrostatic interaction of these charged residues at therotor–stator interface is important for torque generation (Lloyd and Blair,1997; Zhou et al., 1998a). Structure showed that charged residues of FliGare clustered on the prominent ridge of FliGC where two subsets of thecharged residues are aligned as a ‘‘V’’-like shape. FliGM is connected to theFliGC by an a-helix and short linker including well-conserved two consec-utive Gly residues (termed Gly–Gly motif ). This Gly–Gly motif seems to beinvolved in the motion of both domains, acting as a flexible hinge mediatingrelative movements of two connecting domains. These structural featuresof FliG suggest that FliG is likely to have two orientations that bringdifferent subset of charged residues into alignment around the edge of therotor, where they could interact sequentially with charged residues ofthe stator protein MotA. Therefore, such an arrangement of charged resi-dues may be important for the switching of the rotational direction. Yeasttwo-hybrid method was used to investigate the interaction between switchproteins, and revealed two distinct binding sites for FliM in FliG: theEHPQR motif of the FliGM and the hydrophobic patch of the FliGC

(Marykwas et al., 1996). Consistent with this, tryptophan replacements inresidues that participate in these binding sites influenced the binding to FliMand also the assembly of the flagellum (Brown et al., 2007). The hydropho-bic patch in FliGC is positioned adjacent to the Gly–Gly linker and oppositeto charge-bearing ridge. Because the structure of the N-terminal domainhas not been determined yet, it is not clear how FliG binds to FliF.

Ions(Na+ or H+)

4A subunitsFliF

FliM

FliN

EHPQR motif

FliG-N-terminal

region

FliG-C-terminal

regionGly-Gly motif

hydrophobic patch

Conservedcharge residues

Ions(Na+ or H+)

2B subunits

Figure 2.7 The structures of stator and rotor.The stator is formed by theMotA4MotB2or PomA4PomB2 complex. The B subunit has a peptidoglycan-binding motif (pro-truded ball) and the A subunit has large cytoplasmic domain, containing the conservedcharged residues important for flagellar rotation, between second and third transmem-brane segments.The crystal structure of the FliG middle and the C-terminal domain,the FliM middle domain, and the two-third of FliN ofThermatoga maritima are shownusing the PDB data, 1l kv, 2hp7, and 1yab, respectively.The C-terminal region of FliGhas a ridge containing the important charged residues, which are believed to interactwith the charged residues of the cytoplasmic domain ofMotA, for the flagellar rotation.FliN is depicted as a homotetramer composed of dimer-of-dimers. FliG, FliM, and FliNform a switch complex. FliF forms theMS ring of the basal body.



However, as noted above, since FliF–FliG fusion protein is functional, theC-terminal end of FliF and N-terminal end of FliG are likely to be inproximity.

Crystal structure of the middle domain of FliM (residues 44–226, termedFliMM) from T. maritima, which corresponds to two-third of the full-lengthprotein, has been determined (Park et al., 2006) (Fig. 2.7). FliM is involvedin changing of the rotational direction triggered by the binding of CheY-Pto FliM, a chemotactic signaling molecule (Sockett et al., 1992; Welch et al.,1993). The phosphorylation state of CheY is regulated by a kinase (CheA)and phosphatase (CheZ, CheC, or CheX), which are involved in chemo-tactic signaling pathway (Parkinson, 2003). The structure of FliM revealsstructural similarity to these phosphatases. Disulfide cross-linking experi-ments using Cys-substituted FliM variants designed by using the structuralinformation revealed the interface of FliM protein responsible for itsself-association in the C-ring structure. Based on this arrangement of FliMsubunits, about 33–35 copies of FliM subunits form a ring with a diameterof �44 nm, consistent with its rotational symmetry (34-fold) in the C ringobserved in electron cryomicroscopy (Thomas et al., 2006, 1999). CheY-Pbinds to N-terminal portion of FliM, which is not present in the structure.Truncation of the N-terminal 43 residues is required for crystal growth ofFliMM. It is probably disordered when extracted from the C-ring structure.On the other hand, crystal structure of CheY protein in complex with FliMpeptide including N-terminal 16 residues has been reported (Dyer andDahlquist, 2006; Lee et al., 2001). The CheY-FliM peptide structurerevealed the allosteric communication between the phosphorylation siteand the target-binding surface of CheY protein. The C-terminal part ofFliM has been shown to be important for the binding to FliN (Marykwaset al., 1996). A GGXG motif involved in the interaction with FliG exists inthe opposite side of the FliN-binding region (Mathews et al., 1998; Tanget al., 1996; Toker and Macnab, 1997). However, weak electron density inthis region discerns only the first two glycine residues.

The structure of FliN corresponding to the two-third of full-lengthprotein (residues 68–154) from T. maritima (Brown et al., 2005) has beendetermined. FliN, that mostly contributes to form the C-ring structure, isfound to be a tightly intertwisted dimer composed mostly of b-sheet(Fig. 2.7). Structure also revealed a hydrophobic patch, formed by severalwell-conserved hydrophobic residues, on the surface of the FliN dimer.Mutations in these residues in the patch give rise to Fla� or Che� pheno-type, indicating that it is important for the flagellar assembly and the switch-ing. Motility of these mutants is partially restored by the overexpression ofFliI and FliH (Paul et al., 2006), soluble components of export apparatus,suggesting that FliN is involved in the secretion of flagellar proteins. This isnot surprising, since a temperature-sensitive FliN mutant was unable toregrow flagellar filament after shearing it at the restrictive temperature


(Vogler et al., 1991). FliN has been known to involve the export of filamentsubunits or capping proteins. In addition, FliN has a homology to the exportapparatus for virulence factors of pathogenic bacteria (Tang et al., 1995).Therefore, although FliN is involved in switching and motility, it is moredirectly involved in flagellar assembly, probably as a part of flagellum-specific export apparatus. Consistent with this idea, recent biochemicalstudies showed that FliN associates with FliH (McMurry et al., 2006; Paulet al., 2006), and a five-protein complex consisting of FliG, FliM, FliN,FliH, and FliI can be isolated (Gonzalez-Pedrajo et al., 2006). These lines ofevidence suggest that FliN in C ring provides a docking site for exportsubstrate via FliH to efficiently deliver them to the apparatus, and FliN–FliH interaction involves the hydrophobic patch of FliN. Contribution ofFliN to the rotational switching is also involved in the hydrophobic patch.Mutations giving Che� phenotype are mapped around the hydrophobicpatch, and defects could be partially rescued by overexpression of the CheY,suggesting that FliN may contribute to the binding site of CheY-P (Paulet al., 2006). As describe earlier, FliN occupies most part of the C ring, and itis located at the bottom of the ring. Analytical ultracentrifugation of purifiedFliN of T. maritima showed that the FliN exists as a dimer in solution, andFliM and FliN together form the stable FliM1/FliN4 complex (Brown et al.,2005). E. coli FliN exists as a stable homotetramer in solution. These resultsare consistent with the stoichiometry of FliM to FliN in the C ring.Targeted disulfide cross-linking studies of FliN suggested that FliN isorganized in doughnut-shaped tetramers, whose shape is closely matchedfor the bottom of C ring in the reconstructed electron microscopic image(Paul and Blair, 2006).

2.4. Motor complex

The MotA/MotB or PomA/PomB complex that acts as the torque-generating unit exists in the cytoplasmic membrane and assembles aroundthe rotor (Kojima and Blair, 2004a; Yorimitsu and Homma, 2001). MotAand MotB form the torque-generating unit of the Hþ-driven flagellarmotor, whereas PomA and PomB form that of the Naþ-driven flagellarmotor of Vibrio spp. and are orthologues of MotA and MotB (Asai et al.,1997; Dean et al., 1984; Stader et al., 1986) (Figs. 2.7 and 2.8). MotA andPomA have four TMs (Asai et al., 1997; Zhou et al., 1995). A largecytoplasmic loop between the second and third TMs contains conservedcharged residues that have been shown to interact with the conservedcharged residues in the rotor protein FliG, and their electrostatic interac-tions are important for torque generation (discussed also in Section 3.1;Lloyd and Blair, 1997; Zhou and Blair, 1997; Zhou et al., 1998a) (Fig. 2.8).On the other hand, periplasmic loops between first and second TMs(loop1–2) and third and fourth TMs (loop3–4) are very short. MotB and

H+-driven flagellar motorA

MotB(stator)

Periplasm

IMFliG(rotor)

K264

D288

R281

Electrostaticinteraction

CytoplasmMotA(stator)

R297D289

R90

E98

P173 CNN

P222D32

−−

−

−

++

+

+

C

B

(Electrostatic?)interaction

Cytoplasm

Na+-driven flagellar motor

MotX

MotY

PomB(stator)

FliG(rotor)

K284

D308

R301

R317

IM

Periplasm

D309−

−−−

−

+

+

−

++

+

C

NN CP151

P199R88

K89

E96E97 E99 PomA(stator)

D24

Figure 2.8 Charged residues of the putative interaction surface between stator androtor. TheVibrio alginolyticus PomA R88 and E96 correspond to R90 and E98 of theEscherichia coli MotA, and K284, R301, D308, D309, and R317 of V. alginolyticus FliGcorrespond toK264,R281,D288,D289, andR297 ofE. coliFliG, respectively.The asparticacid residue in B subunit (D32 in MotB and D24 in PomB) is believed to be anion-binding site.


PomB have a single TM at its N-terminus (Asai et al., 1997; Chun andParkinson, 1988). This TM contains an absolutely conserved negativelycharged residue (Asp32 in MotB of E. coli and Asp24 in PomB ofV. alginolyticus), which is critical for motor rotation and predicted to bethe ion-binding site in the stator complex (Zhou et al., 1998b). Most ofMotB and PomB proteins are located in the periplasmic space. C-terminalportions of MotB and PomB contain the putative peptidoglycan-binding(PGB) motif that is well conserved among proteins such as OmpA and Pal,


which are outer membrane proteins that interact with the peptidoglycanlayer noncovalently (De Mot and Vanderleyden, 1994; Koebnik, 1995).The MotA/MotB and PomA/PomB complexes are also called the stator,the nonrotating part of the motor. The stator complex forms a heterohex-amer composed of four A subunits (MotA or PomA) and two B subunits(MotB or PomB) (Kojima and Blair, 2004b; Sato and Homma, 2000a,b;Yorimitsu et al., 2004). The MotA/MotB stator functions as a Hþ channeland the PomA/PomB stator functions as an Naþ channel (Blair and Berg,1990; Sato and Homma, 2000a; Stolz and Berg, 1991).

Although structural information is critical to understand the mechanismfor torque generation, currently there are no high-resolution structural dataon the stator complex because its strongly hydrophobic nature hinders toobtain crystals. Instead, arrangements of the TMs of MotA and MotB fromE. coli (18 segments total in a complex) have been investigated by systematicdisulfide cross-linking studies. The results uncovered initial picture of theMotA/MotB stator complex: a symmetric dimer of MotB segments is at thecenter of the complex, and TMs of four MotA molecules are arrangedaround the MotB dimer (Braun and Blair, 2001; Braun et al., 2004). Cysresidues introduced in the third and fourth TMs (TM3 and TM4) of MotAcan form disulfide bridges between those introduced in the single TM ofMotB that contains the proton-accepting Asp residue, suggesting thatproton-conducting channel is formed by these three segments. Arrange-ment of the MotB dimer that fits to the cross-linking results revealed thatthe critical Asp32 residues of two MotB molecules are positioned on theseparate surface of the MotB dimer, so possibly there exist two proton-conducting channels in a MotA4/MotB2 stator complex. This arrangementseems to fit into the PomA/PomB Naþ-conducting stator complex: theTM3 of PomA positions very near the TM of PomB (Yakushi et al., 2004).Phenamil, a potent inhibitor for eukaryotic epithelial sodium channel, alsospecifically inhibits the rotation of Naþ-driven motor (Atsumi et al., 1990).Mutations that allow motor to rotate in the presence of phenamil (Kojimaet al., 1997) were mapped near the cytoplasmic face of the TM3 of PomA(D148Y) and TM of PomB (P16S) ( Jaques et al., 1999; Kojima et al., 1999),and when replaced to Cys, these two residues can form a disulfide bridge(Yakushi et al., 2004). These results suggest that residues in TM3 of PomAand TM of PomB (including critical Asp24) may participate in forming anNaþ-binding site near the cytoplasmic face. Systematic Cys replacement ofthe residues located in the periplasmic loops of PomA revealed that Pro172,which locate in the loop3–4, forms cross-linked dimer under the oxidationcondition, so loop3–4 of two PomAmolecules are positioned side by side ina PomA/PomB complex (Yorimitsu et al., 2000).

The number of the stator complexes assembled around the rotor wasmeasured by various methods. The stepwise increase of the rotation speedthat was dependent on the expression of the stator proteins (Blair and Berg,


1988; Block and Berg, 1984) and the fluorescence intensity change of greenfluorescent protein (GFP)-MotB measured by fluorescence recovery afterphotobleaching (FRAP) method indicate that at least 11 stator complexesare estimated in a single motor (Leake et al., 2006; Reid et al., 2006). Theelectron cryotomography of whole cells of Treponema primitia showed in situstructure of the complete flagellar motor at the 7 nm resolution (Murphyet al., 2006). The image indicates that the stator assembly possessed 16-foldrotational symmetry, within the range of the stator number described above.Recently, the purified PomA/PomB complexes reconstituted into theproteoliposome have been observed by cryo-electron microscopy andrough image of stator was reported (Yonekura et al., 2006). Rod-shapedobjects protruded out from both sides of the lipid bilayer. Its diameter was�20 A, and length of a longer rod and a shorter rod were �70 and 35 A,respectively. The PomA/PomB complex with truncated C-terminus ofPomB, which lacks PGB motif, was also observed, and the longer rod isfound to be the C-terminal domain that contains the PGB motif.

3. Torque Generation

In this chapter, we overview the current knowledge of the mechanismof torque generation, based on genetic and biochemical studies. Manyhypotheses of torque generation have been proposed, and were extensivelyreviewed elsewhere (Berg, 2003; Kojima and Blair, 2004a). It is noteworthythat some interesting motor models were recently published (Schmitt, 2003;Xing et al., 2006).

3.1. Interaction between stator and rotor

It is believed that the rotational force of the flagellar motor is generated bythe interaction between the cytoplasmic loop region of the stator compo-nent, MotA or PomA, and the C-terminal domain of the rotor component,FliG. The rotor–stator interaction is coupled to the Hþ or Naþ flowthrough the stator powered by the electrochemical gradient across thecytoplasmic membrane. Domains of MotA (or PomA) and FliG responsiblefor the rotor–stator interaction contain the conserved charged residuesimportant for the torque generation (Fig. 2.8) (Lloyd and Blair, 1997;Yorimitsu et al., 2002; Zhou and Blair, 1997). In E. coli motor, they wereArg90 and Glu98 of MotA, and Lys264, Arg281, Asp288, Asp289, andArg297 of FliG (Fig. 2.8A). No single residue is critical for rotation, butthey function collectively. Charge neutralization or inversion of theseresidues disrupts motor rotation, but certain combinations of MotA muta-tions with FliG mutations show strong synergism (e.g., MotA-R90A and


FliG R281A) or suppression (e.g., MotA-R90E and FliG D289K), whichsuggests that the charged residues of MotA interact with those of FliG(Zhou et al., 1998a). Therefore, such electrostatic interactions betweenrotor and stator are important for the torque generation. As discussed inSection 2.3, crystal structure of the FliGC revealed that these chargedresidues are clustered on the prominent ridge of the FliGC, where twosubsets of the charged residues (R281/D288/K264 and R281/D289/R297) are aligned as a ‘‘V’’-like shape (Lloyd et al., 1999). Structure andgenetic evidence lead to the rotational switching model that switching mayresult from distinct combination of electrostatic interactions between rotorand stator.

In the case of Naþ-driven motor of V. alginolyticus, these chargedresidues are also conserved in PomA and FliG (R88 and E96 of PomAand K284, R301, D308, D309, and R317 of FliG) (Fig. 2.8B). However,when these residues and neighboring additional three charges (K89, E97,and E99) of PomA were all neutralized, PomA was still functional(Yorimitsu et al., 2002). Likewise, single or all possible combinations ofcharge-neutralizing mutations in five conserved charged residues did notaffect the motility. Inversion of charge in PomA or FliG barely gavenonmotile or slow-motile phenotype (Yorimitsu et al., 2002, 2003).These results suggest that Naþ-driven motor may require additionalcharged residues in PomA and/or FliG for complete electrostatic interac-tion(s). Alternatively, electrostatic interaction between rotor and stator arenot important for the torque generation in V. alginolyticus. It was found thatchimeric FliG protein consisting of the N-terminal two-third fromV. alginolyticus and the C-terminal one-third from E. coli (termed FliGVE)is functional in the fliG strain of V. alginolyticus, and its opposite variant ofchimeric protein (N-terminal two-third from E. coli and C-terminal one-third from V. alginolyticus; FliGEV) is functional in the fliG strain of E. coli(Yorimitsu et al., 2003). Likewise, the MotA/MotB stator from E. coli isfunctional in the DpomAB strain of V. alginolyticus and its motor is driven byproton motive force (PMF) (Asai et al., 2003). Chimeric protein PotB,consisted of the N-terminal TM of PomB from V. alginolyticus and theC-terminal periplasmic segment from E. coli MotB, is functional in theDmotAB strain of E. coli, whose motor is driven by sodium motive force(Asai et al., 2003) (Fig. 2.9A). These results indicate that certain parts of themotor responsible for rotation can be interchangeable between species, andgeneral mechanism of the motor rotation is quite similar regardless of thecoupling ion. To understand the role of charged residues in V. alginolyticusfor motor rotation, conserved charged residues of V. alginolyticus with orwithout mutations were introduced into the hybrid E. colimotor composedof chimeric rotor (FliGEV) and stator proteins (PomA/PotB), and Naþ-driven motility of the cells containing these motor proteins were investi-gated (Yakushi et al., 2006). It was revealed that the charged residues in the

V. alginolyticusE. coli

00 200 400

Speed (Hz)600 800

1000

2000

Tor

que

(pN

nm

) 3000Chimera

V. alginolyticus

E. coli

4000B

A

MotA

MotA PomA

PotB

PomA

PomB

MotX MotY

MotB

MotB

H+ Na+

Na+-driven motor in E. coli

Na+

PomA

PotB

PomA

PomB

Figure 2.9 Hybrid and chimeric motors with the Naþ- and the Hþ-driven compo-nents. (A) The gray and the open part show the regions of theHþ- (MotAorMotB) andthe Naþ-driven components (PomA or PomB), respectively.The horizontal lines showthe outer membrane, peptidoglycan layer, and the cytoplasmic membrane. (B) Thetorque^speed relationships of different flagellarmotors.Measured speeds and estimatedtorques of chimeric motors are represented by circles, the Hþ-driven Escherichia colimotor (squares) using filament drag coefficients (Inoue et al., 2008). Dotted and dashedline are reported torque^speed relationships for Hþ-driven E. coli motor (Chen andBerg, 2000b) andNaþ-drivenVibrio alginolyticusmotor (Sowa et al., 2003), respectively.


V. alginolyticus rotor and stator proteins were found to require for motorrotation when they were engineered in the E. coli motor, although thesynergism and suppression in rotor–stator double mutants were weaker thanthose seen in E. coli. Therefore, basically rotor–stator interaction occurs inthe V. alginolyticus motor in the same way as in E. coli, but the rotor–statorinterface is more robust in V. alginolyticus. Additional charged residues ofPomA may contribute such robustness (Obara et al., 2008), but otherfactors, probably MotX and MotY, may enhance the motor function inV. alginoltyticus. The flagellar motor of S. meliloti rotates only CW direction,and its speed was modulated by the tactic stimuli (Gotz and Schmitt, 1987;Schmitt, 2002). Like E. coli and V. alginolyticus, electrostatic interactions


between conserved charged residues in rotor (R294 and E302 of FliG) andstator (R90, E98, and E150 of MotA) of Sinorhizobium are important fortorque generation. Initially, it was expected that different charge distribu-tion at the rotor–stator interface, due to the absence of several conservedcharged residues in FliG (only two are conserved), might be the basis for thedifferent modes of motor rotation. However, mutational analyses revealedthat unlike E. coli, E150 is essential for torque generation, whereas R90 andE98 are crucial for chemotaxis-controlled modulation of rotation speed(Attmannspacher et al., 2005). Therefore, it was proposed that MotAE150 interacts with FliG R294 to achieve fast rotation, but conformationalchanges in FliG triggered by CheY-P binding to FliM lead to the a newrotor–stator alignment that places E150 of MotA adjacent to D302 of FliG,so that electric repulsion between them might result in lower torque andslower rotation.

Biochemical analysis of the interaction between stator and rotor has notproceeded as compared to the genetic analysis. Only a study has beenreported so far that demonstrated MotA–FliG or MotA–FliM interactionsby pull-down assay (Tang et al., 1996). Then what is the nature of torquegeneration? Although the structure of the stator complex has not beensolved and physical property of the stator complex have been still unclear,systematic mutational studies indicated that MotB Asp32 of E. coli, whichexists in the single TM of MotB, is the only protonatable residue in themotor proteins responsible for the flagellar rotation, suggesting that proton-ation of this critical aspartate in the stator complex triggers conformationalchanges in the stator complex that drive rotor (Zhou et al., 1998b). To testthis idea, protease susceptibility of MotA in complex with MotB wasexamined, and it was revealed that replacement of the critical aspartate(Asp32) of MotB to asparagine or any other small neutral amino acid causeda conformational change in MotA, that could be detected as a change ofprotease susceptibility (Kojima and Blair, 2001). MotA conformation is alsorestricted by the well-conserved prolines located at the cytoplasmic face ofTM3 and 4 (Pro173 and Pro222 in E. coli). In a way that disrupts a-helix,forming b-turn, proline greatly affects secondary and tertiary structures ofthe protein. Mutations in Pro173 and Pro222 abolished or severelyimpaired motility, and mutant proteins exhibit strong dominant-negativeeffect on motility. Therefore, these residues might regulate the conforma-tion of the MotA/MotB complex and/or control conformational changes(Braun et al., 1999; Zhou and Blair, 1997).

3.2. Ion-binding site

As discussed in Section 3.1, an ion-binding site of the stator channel hasbeen believed to be MotB D32 in E. coli and PomB D24 in V. alginolyticus(Yorimitsu and Homma, 2001). Hþ or Naþ associates with the negatively


charged side chain of the acidic amino acid in the stator, although pKa ofcarboxyl group in these residues in the physiological condition has not beenmeasured yet. As described in Section 3.1, when the conserved acidic orbasic amino acid residues in MotA, MotB, FliG, FliM, and FliN, those ofwhich are important for torque generation, were replaced to Ala, all themutants except D32A of MotB still retained motility (Zhou et al., 1998b).Replacement of Asp at position 32 of MotB to various amino acids showedthat all but glutamate at that position were nonfunctional. Therefore, D32of MotB is likely to serve as the proton-binding site in the stator complex.Likewise, PomB D24 probably functions as the binding site for Naþ. It isnoteworthy that an Naþ binds to the functionally critical glutamate residue(E139) in the crystal of NtpK protein, the channel component of anotherNaþ-translocating rotary motor, VOV1-ATPase from Enterococcus hirae(Murata et al., 2005). Recently determined crystal structures ofNaþ-coupledtransporters revealed that not a single but several side chains includingbackbone carbonyl group(s) participate in forming an Naþ-binding site(Hunte et al., 2005; Yamashita et al., 2005). Therefore, in the case of aPomA/PomB complex, there still may exist additional residues involved information of an Naþ-binding site, possibly including D148 of PomA andP16 of PomB, those of which affect phenamil resistancewhenmutated. Also,it is possible that charged residues at the channel entrance or exit or the mainchain carbonyl group(s) of the hydrophobic residues lining in the channelmay play important roles for the selective ion influx (Kojima et al., 2000).

It has been known that cell growth and motility are still normal even ifthe MotA/MotB complex is overexpressed more than 50 times over thewild-type level, suggesting that overproduced complexes are inactive forproton traslocation (Wilson and Macnab, 1988, 1990). When theN-terminal 60 residues of MotB that includes its single TM was fused tothe unrelated polypeptide, consisted of 50 residues encoded by a part ofTetA (MotB60-TetA), its overproduction together with MotA impairs thecell growth (Stolz and Berg, 1991), and this growth impairment wasabolished by the mutation in D32 of MotB (Zhou et al., 1998b). Theseresults led to the proposal that a part of the complex (possibly the periplas-mic domain of MotB) blocks proton flow through the MotA/MotB chan-nel making it inactive. This model is supported by the mutational studydemonstrating that a deletion of the segment just C-terminal to the TM ofMotB (D51–70) or substitution of the residues in that region (I58, Y61, F62,and P52/P65) causes a strong growth impairment when overproduced(Hosking et al., 2006). Detailed analyses of this region brought a modelthat the segment (termed ‘‘plug’’) consists of an amphipathic a-helix and isinserted into the cell membrane parallel with its periplasmic face to interferewith channel formation. Interaction of a MotA/MotB complex with aflagellar basal body triggers movement of the plug from membrane andopening of the proton channel. Therefore, this plug may play a central role


in regulating the open-closed states of the stator channel to prevent prema-ture proton flow.

3.3. Ion specificity

So far, two types of the motor classified by the coupling ion have beenidentified: the Hþ-type and the Naþ-type. Most motile bacteria includingE. coli, S. typhimurium, B. subtilis, R. sphaeroides, and Pseudomonas aeruginosahave Hþ-driven motor and V. alginolyticus and alkalophilic Bacillus haveNaþ-driven motor (Berry and Armitage, 1999; Doyle et al., 2004; Imae andAtsumi, 1989; Yorimitsu and Homma, 2001). It seems that since V. algino-lyticus lives in the sea where Naþ is abundant, it may have evolved to acquireNaþ-conducting activity through the stator complex.

Which part of the stator proteins determines ion specificity? The Hþ-driven stator of the Rhodobacter sphearoides is composed of MotA (RsMotA)and MotB (RsMotB), and RsMotA are found to be remarkably similar tothe Naþ-driven stator of PomA from V. alginolyticus (Asai et al., 1999; Shahand Sockett, 1995). Both of them are composed of 253 amino acids andtheir identity is more than 40% over their entire length. RsMotB hassimilarity only to the transmembrane region of PomB. When RsMotAwas expressed in the pomA strain ofV. alginolyticus, the motor was functionaland driven by Naþ-motive force. On the other hand, expression ofRsMotB together with RsMotA in the pomB strain of V. alginolyticus, cellsare nonmotile (Asai et al., 1999). Therefore, determinants of ion specificityof stator should exist in the PomB (and MotB). A series of chimericproteins, consisting of N-terminal RsMotB and C-terminal PomB (termedMomB) (Asai et al., 2000), was constructed. Some of the MomB constructs,whose entire TMwas derived fromRsMotB, functioned as Naþ-type statorwhen coexpressed with RsMotA in the DpomAB strain of V. alginolyticus.In this case, four proteins RsMotA, MomB, MotX, and MotY are involvedin the Naþ-driven rotation. Interestingly, a MomB construct, whose junc-tion is between F33 and V34 of PomB, functions better as Liþ-driven motorthan as Naþ-driven motor. Later, it was found that a chimeric protein ofPotB, a fusion of N-terminal TM of PomB and the C-terminal periplasmicsegment of E. coli MotB, functions as an Naþ-type stator in E. coli onlywhen coexpressed with PomA (Asai et al., 2003). In this case, just twoproteins PomA and PotB are required for Naþ-driven rotation (Fig. 2.9A).These lines of evidence indicate that cytoplasmic and transmembranedomains of PomA/PomB complex are sufficient for Naþ-driven motility,and periplasmic C-terminal part of B subunit (MotB or PomB) determinesthe requirement of MotX and MotY for function. Probably, the size ofchannel pore is varied in these chimeric stator complexes, altering the ionspecificity. More precise descriptions of ion specificity in the stator will beprovided by the high-resolution atomic model of the stator complex.


3.4. Assembly of functional motor

Whereas the assembly mechanism of the axial flagellar structure is now wellunderstood (see Section 1.3), themechanismof stator assembly is still not clear.Early experiments showed that controlled expression of motA ormotB in theirdefective strains demonstrated stepwise restoration of the rotation rate withequal speed increment that saturated at eight steps, suggesting that eightindependently functioning MotA/MotB stator units are incorporated intothe wild-type motor (Blair and Berg, 1988; Block and Berg, 1984). Recentstudies with higher resolution reported that as much as 11 stator units can beincorporated (Reid et al., 2006). This number is in proximity to the observedparticles that surround the rotor by freeze-fracture images (Khan et al., 1988).MotB has a putative PGBmotif that is well conserved among proteins such asOmpA and Pal, which are outer membrane proteins that interact with thepeptidoglycan layer noncovalently (De Mot and Vanderleyden, 1994;Koebnik, 1995). The PGB motif of MotB is believed to associate with thepeptidoglycan layer to anchor the MotA/MotB stator complex around therotor. It is not clear how andwhenMotA/MotB stator complexes are targetedand anchored at the appropriate position around the rotor.

In V. alginolyticus and V. parahaemolyticus, MotX and MotY have beenidentified as essential components for the rotation of Naþ-driven polarflagellar motors (McCarter, 1994a,b; Okabe et al., 2001; Okunishi et al.,1996). As described in Section 2.1, they attach to the basal body to form a ringstructure (termed T ring) that can be observed beneath the LP ring byelectron microscopy (Terashima et al., 2006). Biochemical studies haveshown that MotX directly interacts with MotY, and affects membranelocalization of the PomA/PomB complex and of the PomB alone,suggesting an interaction between MotX and PomB (Okabe et al., 2005).In addition, whenMotX orMotY is absent, GFP-fused PomAorGFP-fusedPomB in complex with their partner subunit does not localize at theflagellated cell pole, suggesting that MotX and MotY in the T ring areinvolved in the incorporation of PomA/PomB complex into the flagellarmotor (Terashima et al., 2006). Like MotB and PomB, MotY possessespeptidoglycan-binding motif and a recently solved MotY structure showedremarkably similar structure to the Pal and RmpM (OmpA homologue),well-known peptidoglycan-binding protein (Kojima et al., 2008). Detailedfunctions of MotX and MotY are still unclear, but they are speculated to beinvolved in ion specificity, or in extremely rapid rotation (1100 Hz)approximately four times faster than that of E. coli (300 Hz).

In S. meliloti, MotC andMotE have shown to be involved in the motilityand regulation of the rotation speed of the flagllar motor (Eggenhofer et al.,2004; Platzer et al., 1997). MotC is the periplasmic protein and regulates therotation speed by acting onMotB. MotE is the specific chaperone for MotCand controls the rotation speed indirectly by regulating amount of the


MotC proteins in periplasm. These proteins do not show similarity toMotX and MotY. In some bacteria, two kinds of stator might be assembledto a single flagellar base. Pseudomonas aeruginosa, which has a single polarflagellum, has dual sets of motA and motB genes, motAB and motCD, as wellas another gene, motY. All these five genes contribute to Hþ-driven motility(Doyle et al., 2004). Function of MotA/MotB stator requires MotY, andthese three proteins are important for the surface swarming. On the otherhand, MotC/MotD does not require MotY for its function, and they areimportant for the swimming in liquid. Furthermore, noncognate pairs likeMotA/MotD and MotB/MotC can work together to generate torque:mutants that contain motA/motD or motB/motC double mutations still retainmotility. Therefore, Pseudomonas cells seem to choose two types of statorsfor motility depending on the surrounding environment (swimming orsurface swarming) (Doyle et al., 2004; Toutain et al., 2005). Similarly,when swarming on the surface, V. parahaemolyticus as well as V. alginolyticuscells induced multiple lateral flagella that are driven by proton-type motor(Atsumi et al., 1992; McCarter et al., 1988). The stator for lateral flagellarmotors contains LafT and LafU, orthologues of MotA and MotB, andsomehow its function requires a second set of MotY, MotYL (Stewart andMcCarter, 2003). The reason why MotYL is necessary for rotation of thelateral flagellar motor is not clear.

Another example is the motor of B. subtilis: it has an Naþ-driven MotP/MotS stator and a proton-driven MotA/MotB stator, but a single set offlagellar rotor proteins (Ito et al., 2004). Like the hybrid stator of P.aeruginosa, hybrid stators like MotP/MotB and MotA/MotS are functionalin B. subtilis and ion specificities of these motors depend on the B subunit(MotA/MotS for Naþ-type and MotP/MotB for Hþ-type) (Ito et al.,2005). These results are consistent with the observation of hybrid motorswith chimeric proteins using stator proteins of V. alginolyticus and R.sphaeroides. Since cells carrying MotP/MotS and MotP/MotB stator canswim faster in liquid with high viscosity, MotP function is suggested to beimportant for the optimal function in elevated viscosity. In order to achieveoptimal behavior in variable environments, these bacteria seem to evolve touse distinct sets of stator complexes or have additional components besidesconventional stator proteins to fully exert motor rotation.

4. Molecular Physiology of Motor

4.1. Torque–speed relationship

To understand the mechanism of motor rotation, we need to know itsbasic properties: the power input and output, and their relationships.Measuring these properties of the motor involves technical difficulties, so


understanding of the motor physiology has been accompanied by technicaldevelopment in the measurement system. Although the studies of powerinput for a single motor have still been hindered by such difficulties, one canmeasure the power output as the ‘‘rotation’’ by using a quite simple methodthat has been employed since the discovery of flagellar rotation. Cells areattached to a coverslip by a single flagellar filament and the rotation of eachsingle motor is monitored by the resulting rotation of the cell body(Silverman and Simon, 1974). Such tethered cells rotate slowly (around10 Hz) because of the large viscous load. Measurement of rotation of lightlyloaded motors in swimming cells was achieved by a light scattering methodunder the microscope (Lowe et al., 1987), and values obtained by these twomeasurements were used for the initial estimation of a torque–speed rela-tionship, an essential feature that characterizes the flagellar motor. Theresults showed a linear relationships: that as the speed increases, the torquelinearly decreases.

To cover a wide range of the speed by the measurement of a singlemotor, the electrorotation method was developed (Berg and Turner, 1993;Iwazawa et al., 1993). In this technique, a tethered cell was spun under arotating electric field that enabled motors to rotate up to several hundreds ofHertz. Measurement of the motor rotation has been carried out by thismethod, and revealed characteristic feature of the torque–speed relation-ships for the flagellar motor: for forward rotation, motor torque remainedapproximately constant up to speeds of about 60% of the zero-torque speed,then the torque dropped linearly with speed, crossed zero, and reached aminimum (Berg and Turner, 1993). In yet another approach, the cell bodywas fixed to the glass surface and a polystyrene bead was attached to a stub ofone of its flagellar filaments (Ryu et al., 2000). By using beads of differentsizes and changing the viscosity of external medium, one can measure motortorque under a wide range of speeds. The results confirmed the previousmeasurement by electrorotation method, showing the essentially constanttorque and then linear decline as the speed increases (Chen and Berg,2000b). The latter method allows us to measure the rotation at variousconditions, such as different temperatures and using D2O instead of H2Ofor the effect of solvent isotope (Chen and Berg, 2000a). The results showedthat in the low-speed regime, torque was independent of temperature, andsolvent isotope effects were relatively small.

In the high-speed regime, torque was strongly dependent on tempera-ture, as seen by a downward shift in the ‘‘knee’’ value, the region of thetransition from constant torque to declining torque, at lower temperature.Also, solvent isotope effects were large. These results were consistent withprevious studies for artificially energized cells of Streptococcus (Khan andBerg, 1983; Manson et al., 1980). Therefore, torque–speed relationship ofthe flagellar motor indicates that at low speeds, the motor operates nearthermodynamic equilibrium, where rates of movement of the internal


mechanical components or translocation of protons are not rate-limiting,and that at high speeds, the rate-limiting step is proton transfer event, whichresults in the decline in torque at high speed. As discussed by Berry andBerg, a torque–speed relationship with this shape can be interpreted in thecontext of a simple three-state kinetic model and it suggests a rotationmechanism with a power stroke, in which motor rotation and dissipationof the energy available from proton transit occur synchronously (Berry andBerg, 1999). Fung and Berg (1995) found that when the motor operates inthe low-speed regime near stall, its speed is proportional to the voltageapplied across the inner cell membrane. Moreover, Gabel and Berg (2003)reported a linear relationship between speed (0–270 Hz) and PMF for lightload on the motor. If we assume tightly coupled motor in which thetranslocation of a fixed number of protons drives each rotation, the linearspeed–PMF relation is indicative of a simple voltage-gated proton channel.

Comparative studies have been reported for the Naþ-driven polarflagellar motor of V. alginolyticus (Sowa et al., 2003). It has been shownthat the torque–speed curve had the same shape as those of the wild-typeE. coli motor and the hybrid motor with PomA/PotB chimeric statorin E. coli with full expression of the stator proteins: the torque is approxi-mately constant (at �2200 pN nm) from stall up to a ‘‘knee’’ speed of�420 Hz, and then falls linearly with speed, extrapolating to zero torque at�910 Hz (Inoue et al., 2008) (Fig. 2.9B). The overall shape of the torque–speed curve is quite similar to the Hþ-driven E. coli motor describedearlier, but the effect of the concentrations of the coupling ion ontorque–speed curves of V. alginolyticus was different from those of E. coli.The curves for E. coli did not change even when the external environmenthad a pH value in the range of 4.7–8.8. However, when external NaClconcentrations were changed, the generated torque was changed over awide range of speeds for Vibrio motor. This difference seems to be derivedfrom not the mechanism of flagellar motors but the cell homeostasis whenthe concentration of the external coupling ion is varied. The Naþ-drivenmotor is likely to generate torque using basically the same mechanism as theHþ-driven motor.

4.2. Steps in rotation of motor

The rotation of the motor occurs by a sequence of discrete molecularevents, presumably including the stages of ion binding in the stator followedby conformational changes in the stator that drive the rotor. Therefore, onecan imagine that the motor behaves like a stepping motor. Since thediscovery of flagellar rotation, such a step has been investigated but untilrecently it has not clearly demonstrated. The difficulty in observing rota-tional steps is due to the presence of an elastic linkage, the hook, betweenthe tethered filament and the cell body, which smoothes the observed


rotation, and is also due to multiple independently functioning torque-generating units (stator complexes) per motor. One way to avoid thisdifficulty is to examine variations in rotation period. Such a stochasticanalysis by assuming that intervals between steps follow a Poisson distribu-tion led to an estimation of about 400 steps per revolution (Samuel andBerg, 1995). A similar analysis was done with motors containing only one ora few torque-generating units, and showed that the individual units stepindependently (Samuel and Berg, 1996).

Recently, direct observation of steps in the motor has been achieved bymaking a motor rotate as slow as possible in the work of Sowa et al (2005).The hybrid E. coli motor equipped with chimeric Naþ-driven stator com-plex PomA/PotB was used under the expression control of induciblepromoter. Therefore, it was able to establish a motor with a small initialnumber of stator units, possibly one, by growing cells with low inducerlevel. Furthermore, by using the Naþ-type motor, the rotation rate can bereduced by lowering external Naþ concentration. Motor rotation wasdetected by back-focal-plane (BFP) interferometry of 500-nm diameterbeads attached to spontaneously sticky flagellar filament stub or by high-speed video recording of 200-nm diameter fluorescent beads attached in thesame way. By using these experimental setups, the stepping rotations wereobserved at speeds below 7 and 40 Hz in BFP and fluorescence experi-ments, respectively. Interestingly, both backward and forward steps wereobserved at all speeds in both experiments, with higher probability at lowerspeeds. In this experiment, a cheY strain was used so that motor never rotatesCW, thus backward steps represent microscopic reversibility rather thanmotor switching. Similar steps have been detected in the ATP-drivenmolecular motors. The step size was analyzed from these observations,and it was revealed that there is a peak at 13.7� (26 per revolution) and�10.3� (35 per revolution) for forward and backward steps, respectively.Stepping motion in the ATP-driven molecular motors reflects both thediscrete nature of the input energy and the periodicity of the ‘‘track’’ alongwhich the motor runs (Mehta et al., 1999; Schnitzer and Block, 1997;Yasuda et al., 1998). Likewise, observation of 26 steps per revolution inthe flagellar motor is consistent with the periodicity of the ring of FliGprotein (see Section 2.3), suggesting that the motor steps along each FliGmolecule in the rotor. Then how many ions can be translocated during onestep? Because the simultaneous measurement of a single motor rotation andthe ion translocation through the stator complex has not yet been estab-lished, we can only estimate it from the published parameters. It wasreported that a wild-type E. coli cell with a PMF of about 150 mV drivesa 1 mm bead with an estimated 280 pN nm per unit (Ryu et al., 2000),suggesting that maximal step size for one ion in one unit would be 5� if allthe input energy was consumed for the torque generation. Thus, there maybe smaller substeps, or alternatively mechanical step may be coupled to


several ion translocations. As discussed in Section 2.4, the structural modelfor the stator complex proposed by the disulfide cross-linking studysuggested two distinct channels per stator complex (Braun and Blair,2001; Braun et al., 2004), indicative of the latter possibility. In any case,the atomic model of the stator complex and establishment of simultaneousmeasurement system to investigate input–output relationships will beawaited for solving this problem.

4.3. Fluorescent imaging of motor

Recent development of single-molecule measurements of enzyme activitiesby using fluorescent microscopic methods makes it possible to reveal muchabout fundamental mechanisms of these protein machines. Toward thesimultaneous measurement of the single motor rotation and ion fluxthrough the stator complex, several attempts of fluorescent imaging ofmotor components have been undertaken. Our laboratory has used GFPfor such an imaging in vivo (Fukuoka et al., 2007, 2005). GFP was fused tothe stator proteins, PomA and PomB of V. alginolyticus, and their behaviorsin vivo were observed under the fluorescent microscope (Fukuoka et al.,2005). By using this system, first, it was confirmed that GFP-PomA or GFP-PomB proteins were actually localized at the flagellated cell pole of Vibrio.Their polar localization requires the partner subunit (PomB or PomA) andthe C-terminal domain of PomB that contains PGB motif required foranchoring the stator complexes to the motor. The polar localization ofGFP-fused stator proteins was not observed in the absence of the polarflagellum. Finally, cells that express the PomA/GFP-PomB stator exhibitedmotility, although it was significantly reduced as compared with wild-typestrain. Therefore, this is the initial success in imaging the functional statorproteins in vivo. Since the PomA/PomB stator complex assembles into thesingle polar flagellum of Vibrio, assembly of the stator complex at theflagellated cell pole can be observed as the polar localization of the GFP-fused stator proteins.

When we observed their behavior in the absence of Naþ, the couplingion for polar flagellar motor of Vibrio, the GFP-fused stator proteins werenot localized at the flagellated cell pole (Fukuoka et al., submitted forpublication). This effect appears to be Naþ-dependent and reversible,since addition of Naþ in the medium restores the polar localization of theGFP-fused stators. The mutations of PomB D24, which is the binding sitefor Naþ, affected the localization. Therefore, stator assembly seems to beregulated by the energy source of the motor, the Na-motive force. Thisfinding also suggested more dynamic behavior of the stator proteins thanpreviously expected. These results are consistent with the previous reports


(Armitage and Evans, 1985; Evans and Armitage, 1985) and the recentreport by Leake et al. (2006), demonstrating simultaneous measurement ofthe single motor rotation and the number and dynamics of GFP-fusedMotB molecules in the motor in single molecule level by total internalreflection fluorescence (TIRF) microscopy. They tethered cells expressingGFP-MotB, and observed them under the TIRF or brightfield microscopy.TIRF images of tethered cells showed spots at the center of the cell rotationmeasured from brightfield images, indicating functional assembly of thestator consisting of GFP-MotB and MotA. Then they carried out stepwisephotobleaching of single GFP-MotB in the motor. Counting fluorophoresrevealed that each motor contains around 22 copies of GFP-MotB, consis-tent with 11 stators reported by the recent resurrection experiment. Theyalso reported by the analyses using FRAP and fluorescence loss in photo-bleaching methods that (1) there is a membrane pool of �200 GFP-MotBmolecules diffusing at �0.008 mm2 s�1, and (2) turnover of GFP-MotBbetween the membrane pool and motor was observed with a rate constantof the order of 0.04 s�1. Therefore, the static image of the ‘‘stator’’ is in needof change, and we need to understand its function under the considerationof more dynamic behavior. The stator complex appears to be transientlyanchored around the rotor to generate torque, and rapidly exchanged whenit worn or by responding to the change in energetic environment in the cell.

Fluorescent imaging of the rotor protein has also been attempted, andwe were able to visualize functional GFP-FliG molecules in the center ofthe rotating tethered cell (Fukuoka et al., 2007). We are expecting thatappropriate combination of functional GFP-fused stator and rotor proteinin the motor may enable us to carry out measurement of FRET betweenstator and rotor components to demonstrate real-time imaging of rotor–stator interaction in vivo.

Imaging of the energy input into the motor is another big challenge.Recently, using a fluorescent probe for Naþ, Sodium Green, intracellularNaþ concentration in single E. coli cells has been measured (Lo et al., 2006).This method requires low-light electron-multiplying charge-coupleddevice camera and laser fluorescence microscopy, and makes it possible tomeasure intracellular Naþ of a series of 50 single cells, without any detect-able effect on the flagellar motor. The values obtained are consistent withthe ones measured by other methods. Likewise, fluorescence technique tomeasure membrane potential in single cell has been established by using thedye tetramethylrhodamine methyl ester (Lo et al., 2007). These fluorescencetechniques to measure input energy (membrane potential and ion gradient)together with the rotation analysis in the single molecule level finally openthe new era for motor physiology that would make it possible to visualizeinput–output relationships in real time and in vivo.


5. Conclusion

In addition to the proton-driven motor, the Na+-driven motor hasbeen studied extensively and many important data have accumulated. Usingthese insights, a genetic manipulation of the Na+-driven E. coli hybridmotor with chimeric stator led us to a recent breakthrough to observedirectly the steps in rotation of a single motor, the basic process of themotor. From now on, we can expect to elucidate the rotation mechanismby discussing the input and output relations of the energy during a singlestep in a rotation. Moreover, the technology of single-molecule fluorescentobservation has been introduced, and it will be able to visualize a dynamicinteraction between rotor and stator. To understand the mechanism ofenergy conversion that changes the ion flux into the mechanical power,the crystal structures of the membrane motor proteins are also needed. Wewould like to learn the biological nature from the tiny nanomachine of thebacterial flagella.

REFERENCES

Aizawa, S., Dean, G. E., Jones, C. J., Macnab, R. M., and Yamaguchi, S. (1985). Purifica-tion and characterization of the flagellar hook-basal body complex of Salmonella typhi-murium. J. Bacteriol. 161, 836–849.

Akiba, T., Yoshimura, H., and Namba, K. (1991). Monolayer crystallization of flagellar L-Prings by sequential addition and depletion of lipid. Science 252, 1544–1546.

Aldridge, P., Karlinsey, J. E., Becker, E., Chevance, F. F., and Hughes, K. T. (2006). Flkprevents premature secretion of the anti-sigma factor FlgM into the periplasm. Mol.Microbiol. 60, 630–643.

Armitage, J. P. (1999). Bacterial tactic responses. Adv. Microb. Physiol. 41, 229–289.Armitage, J. P., and Evans, M. C. W. (1985). Control of the proton-motive force in

Rhodopseudomonas sphaeroides in the light and dark and its effect on the initiation offlagellar rotation. Biochimica Biophysica Acta 806, 42–45.

Armitage, J. P., and Macnab, R. M. (1987). Unidirectional, intermittent rotation of theflagellum of Rhodobacter sphaeroides. J. Bacteriol. 169, 514–518.

Asai, Y., Kojima, S., Kato, H., Nishioka, N., Kawagishi, I., and Homma, M. (1997).Putative channel components for the fast-rotating sodium-driven flagellar motor of amarine bacterium. J. Bacteriol. 179, 5104–5110.

Asai, Y., Kawagishi, I., Sockett, R. E., and Homma, M. (1999). Hybrid motor with Hþ- andNaþ-driven components can rotate Vibrio polar flagella by using sodium ions. J. Bacteriol.181, 6332–6338.

Asai, Y., Kawagishi, I., Sockett, R. E., and Homma, M. (2000). Coupling ion specificity ofchimeras between Hþ- and Naþ-driven motor proteins, MotB and PomB, in Vibrio polarflagella. EMBO J. 19, 3639–3648.

Asai, Y., Yakushi, T., Kawagishi, I., and Homma, M. (2003). Ion-coupling determinants ofNaþ-driven and Hþ-driven flagellar motors. J. Mol. Biol. 327, 453–463.

Asakura, S. (1970). Polymerization of flagellin and polymorphism of flagella. Adv. Biophys. 1,99–155.


Atsumi, T., Sugiyama, S., Cragoe, E. J., Jr., and Imae, Y. (1990). Specific inhibition of theNaþ-driven flagellar motors of alkalophilic Bacillus strains by the amiloride analogphenamil. J. Bacteriol. 172, 1634–1639.

Atsumi, T., McCarter, L., and Imae, Y. (1992). Polar and lateral flagellar motors of marineVibrio are driven by different ion-motive forces. Nature 355, 182–184.

Attmannspacher, U., Scharf, B., and Schmitt, R. (2005). Control of speed modulation(chemokinesis) in the unidirectional rotary motor of Sinorhizobium meliloti.Mol. Microbiol.56, 708–718.

Bange, G., Petzold, G., Wild, K., Parlitz, R. O., and Sinning, I. (2007). The crystal structureof the third signal-recognition particle GTPase FlhF reveals a homodimer with boundGTP. Proc. Natl. Acad. Sci. USA 104, 13621–13625.

Bennett, J. C., Thomas, J., Fraser, G. M., and Hughes, C. (2001). Substrate complexes anddomain organization of the Salmonella flagellar export chaperones FlgN and FliT. Mol.Microbiol. 39, 781–791.

Berg, H. C. (2003). The rotary motor of bacterial flagella. Annu. Rev. Biochem. 72, 19–54.Berg, H. C., and Anderson, R. A. (1973). Bacteria swim by rotating their flagellar filaments.

Nature 245, 380–382.Berg, H. C., and Turner, L. (1993). Torque generated by the flagellar motor of Escherichia

coli. Biophys. J. 65, 2201–2216.Berry, R. M., and Armitage, J. P. (1999). The bacterial flagella motor. Adv. Microb. Physiol.

41, 291–337.Berry, R. M., and Berg, H. C. (1999). Torque generated by the flagellar motor of Escherichia

coli while driven backward. Biophys. J. 76, 580–587.Blair, D. F., and Berg, H. C. (1988). Restoration of torque in defective flagellar motors.

Science 242, 1678–1681.Blair, D. F., and Berg, H. C. (1990). The MotA protein of E. coli is a proton-conducting

component of the flagellar motor. Cell 60, 439–449.Block, S. M., and Berg, H. C. (1984). Successive incorporation of force-generating units in

the bacterial rotary motor. Nature 309, 470–472.Braun, T. F., and Blair, D. F. (2001). Targeted disulfide cross-linking of the MotB protein of

Escherichia coli: Evidence for two Hþ channels in the stator complex. Biochemistry 40,13051–13059.

Braun, T. F., Poulson, S., Gully, J. B., Empey, J. C., Van Way, S., Putnam, A., andBlair, D. F. (1999). Function of proline residues of MotA in torque generation by theflagellar motor of Escherichia coli. J. Bacteriol. 181, 3542–3551.

Braun, T. F., Al-Mawsawi, L. Q., Kojima, S., and Blair, D. F. (2004). Arrangement of coremembrane segments in the MotA/MotB proton-channel complex of Escherichia coli.Biochemistry 43, 35–45.

Brown, P. N., Hill, C. P., and Blair, D. F. (2002). Crystal structure of the middle andC-terminal domains of the flagellar rotor protein FliG. EMBO J. 21, 3225–3234.

Brown, P. N., Mathews, M. A., Joss, L. A., Hill, C. P., and Blair, D. F. (2005). Crystalstructure of the flagellar rotor protein FliN from Thermotoga maritima. J. Bacteriol. 187,2890–2902.

Brown, P. N., Terrazas, M., Paul, K., and Blair, D. F. (2007). Mutational analysis of theflagellar protein FliG: Sites of interaction with FliM and implications for organization ofthe switch complex. J. Bacteriol. 189, 305–312.

Calladine, C. R. (1978). Change of wave form in bacterial flagella: The role of mechanics atthe molecular level. J. Mol. Biol. 118, 457–479.

Chen, X., and Berg, H. C. (2000a). Solvent-isotope and pH effects on flagellar rotation inEscherichia coli. Biophys. J. 78, 2280–2284.

Chen, X., and Berg, H. C. (2000b). Torque-speed relationship of the flagellar rotary motorof Escherichia coli. Biophys. J. 78, 1036–1041.


Chevance, F. F., Takahashi, N., Karlinsey, J. E., Gnerer, J., Hirano, T., Samudrala, R.,Aizawa, S., and Hughes, K. T. (2007). The mechanism of outer membrane penetrationby the eubacterial flagellum and implications for spirochete evolution. Genes Dev. 21,2326–2335.

Chilcott, G. S., and Hughes, K. T. (2000). Coupling of flagellar gene expression to flagellarassembly in Salmonella enterica serovar typhimurium and Escherichia coli.Microbiol. Mol. Biol.Rev. 64, 694–708.

Chun, S. Y., and Parkinson, J. S. (1988). Bacterial motility: Membrane topology of theEscherichia coli MotB protein. Science 239, 276–278.

Claret, L., Calder, S. R., Higgins, M., and Hughes, C. (2003). Oligomerization andactivation of the FliI ATPase central to bacterial flagellum assembly. Mol. Microbiol. 48,1349–1355.

Cluzel, P., Surette, M., and Leibler, S. (2000). An ultrasensitive bacterial motor revealed bymonitoring signaling proteins in single cells. Science 287, 1652–1655.

Cordell, S. C., and Lowe, J. (2001). Crystal structure of the bacterial cell division regulatorMinD. FEBS Lett. 492, 160–165.

Cornelis, G. R. (2006). The type III secretion injectisome. Nat. Rev. Microbiol. 4, 811–825.Dailey, F. E., and Berg, H. C. (1993). Mutants in disulfide bond formation that disrupt

flagellar assembly in Escherichia coli. Proc. Natl. Acad. Sci. USA 90, 1043–1047.De Mot, R., and Vanderleyden, J. (1994). The C-terminal sequence conservation between

OmpA-related outer membrane proteins and MotB suggests a common function in bothGram-positive and Gram-negative bacteria, possibly in the interaction of these domainswith peptidoglycan. Mol. Microbiol. 12, 333–334.

Dean, G. D., Macnab, R. M., Stader, J., Matsumura, P., and Burks, C. (1984). Genesequence and predicted amino acid sequence of the motA protein, a membrane-associated protein required for flagellar rotation in Escherichia coli. J. Bacteriol. 159,991–999.

Doyle, T. B., Hawkins, A. C., and McCarter, L. L. (2004). The complex flagellar torquegenerator of Pseudomonas aeruginosa. J. Bacteriol. 186, 6341–6350.

Dyer, C. M., and Dahlquist, F. W. (2006). Switched or not? The structure of unpho-sphorylated CheY bound to the N-terminus of FliM. J. Bacteriol. 188, 7354–7363.

Eggenhofer, E., Haslbeck, M., and Scharf, B. (2004). MotE serves as a new chaperonespecific for the periplasmic motility protein, MotC, in Sinorhizobium meliloti. Mol. Micro-biol. 52, 701–712.

Evans, M. C. W., and Armitage, J. P. (1985). Initiation of flagellar rotation in Rhodopsudo-monas sphaeroides. FEBS Lett. 186, 93–97.

Fahrner, K. A., Block, S. M., Krishnaswamy, S., Parkinson, J. S., and Berg, H. C. (1994). Amutant hook-associated protein (HAP3) facilitates torsionally induced transformations ofthe flagellar filament of Escherichia coli. J. Mol. Biol. 238, 173–186.

Fan, F., and Macnab, R. M. (1996). Enzymatic characterization of FliI—An ATPaseinvolved in flagellar assembly in Salmonella typhimurium. J. Biol. Chem. 271,31981–31988.

Fan, F., Ohnishi, K., Francis, N. R., and Macnab, R. M. (1997). The FliP and FliR proteinsof Salmonella typhimurium, putative components of the type III flagellar export apparatus,are located in the flagellar basal body. Mol. Microbiol. 26, 1035–1046.

Ferris, H. U., and Minamino, T. (2006). Flipping the switch: Bringing order to flagellarassembly. Trends Microbiol. 14, 519–526.

Focia, P. J., Shepotinovskaya, I. V., Seidler, J. A., and Freymann, D. M. (2004). Hetero-dimeric GTPase core of the SRP targeting complex. Science 303, 373–377.

Francis, N. R., Irikura, V. M., Yamaguchi, S., DeRosier, D. J., and Macnab, R. M. (1992).Localization of the Salmonella typhimurium flagellar switch protein FliG to the cytoplasmicM-ring face of the basal body. Proc. Natl. Acad. Sci. USA 89, 6304–6308.


Francis, N. R., Sosinsky, G. E., Thomas, D., and DeRosier, D. J. (1994). Isolation,characterization and structure of bacterial flagellar motors containing the switch complex.J. Mol. Biol. 235, 1261–1270.

Fraser, G. M., Bennett, J. C., and Hughes, C. (1999). Substrate-specific binding of hook-associated proteins by FlgN and FliT, putative chaperones for flagellum assembly. Mol.Microbiol. 32, 569–580.

Fraser, G. M., Gonzalez-Pedrajo, B., Tame, J. R., and Macnab, R. M. (2003). Interactionsof FliJ with the Salmonella type III flagellar export apparatus. J. Bacteriol. 185, 5546–5554.

Fukuoka, H., Yakushi, T., Kusumoto, A., and Homma, M. (2005). Assembly of motorproteins, PomA and PomB, in the Naþ-driven stator of the flagellar motor. J. Mol. Biol.351, 707–717.

Fukuoka, H., Sowa, Y., Kojima, S., Ishijima, A., and Homma, M. (2007). Visualization offunctional rotor proteins of the bacterial flagellar motor in the cell membrane. J. Mol.Biol. 367, 692–701.

Fung, D. C., and Berg, H. C. (1995). Powering the flagellar motor of Escherichia coli with anexternal voltage source. Nature 375, 809–812.

Gabel, C. V., and Berg, H. C. (2003). The speed of the flagellar rotary motor of Escherichiacoli varies linearly with protonmotive force. Proc. Natl. Acad. Sci. USA 100, 8748–8751.

Gillen, K., and Hughes, K. T. (1991). Negative regulatory loci coupling flagellin synthesis toflagellar assembly in Salmonella typhimurium. J. Bacteriol. 173, 2301–2310.

Gonzalez-Pedrajo, B., Minamino, T., Kihara, M., and Namba, K. (2006). Interactionsbetween C ring proteins and export apparatus components: A possible mechanism forfacilitating type III protein export. Mol. Microbiol. 60, 984–998.

Gotz, R., and Schmitt, R. (1987). Rhizobium meliloti swims by unidirectional, intermittentrotation of right-handed flagellar helices. J. Bacteriol. 169, 3146–3150.

Hirano, T., Yamaguchi, S., Oosawa, K., and Aizawa, S. I. (1994). Roles of FliK and FlhB indetermination of flagellar hook length in Salmonella typhimurium. J. Bacteriol. 176,5439–5449.

Hizukuri, Y., Yakushi, T., Kawagishi, I., and Homma, M. (2006). Role of the intramolec-ular disulfide bond in FlgI, the flagellar P-ring component of Escherichia coli. J. Bacteriol.188, 4190–4197.

Homma, M., and Iino, T. (1985). Locations of hook-associated proteins in flagellar struc-tures of Salmonella typhimurium. J. Bacteriol. 162, 183–189.

Homma, M., Fujita, H., Yamaguchi, S., and Iino, T. (1984). Excretion of unassembledflagellin by Salmonella typhimurium mutants deficient in the hook-associated proteins.J. Bacteriol. 159, 1056–1059.

Homma, M., Iino, T., Kutsukake, K., and Yamaguchi, S. (1986). In vitro reconstitution offlagellar filaments onto hooks of filamentless mutants of Salmonella typhimurium byaddition of hook-associated proteins. Proc. Natl. Acad. Sci. USA 83, 6169–6173.

Homma, M., Komeda, Y., Iino, T., and Macnab, R. M. (1987). The flaFIX gene product ofSalmonella typhimurium is a flagellar basal body component with a signal peptide forexport. J. Bacteriol. 169, 1493–1498.

Homma, M., Kutsukake, K., Hasebe, M., Iino, T., and Macnab, R. M. (1990). FlgB, FlgC,FlgF and FlgG. A family of structurally related proteins in the flagellar basal body ofSalmonella typhimurium. J. Mol. Biol. 211, 465–477.

Hosking, E. R., Vogt, C., Bakker, E. P., and Manson, M. D. (2006). The Escherichia coliMotAB proton channel unplugged. J. Mol. Biol. 364, 921–937.

Hueck, C. J. (1998). Type III protein secretion systems in bacterial pathogens of animals andplants. Microbiol. Mol. Biol. Rev. 62, 379–433.

Hughes, K. T., Gillen, K. L., Semon, M. J., and Karlinsey, J. E. (1993). Sensing structuralintermediates in bacterial flagellar assembly by export of a negative regulator. Science 262,1277–1280.


Hunte, C., Screpanti, E., Venturi, M., Rimon, A., Padan, E., and Michel, H. (2005).Structure of a Naþ/Hþ antiporter and insights into mechanism of action and regulationby pH. Nature 435, 1197–1202.

Iino, T. (1969). Genetics and chemistry of bacterial flagella. Bacteriol. Rev. 33, 454–475.Ikeda, T., Asakura, S., and Kamiya, R. (1985). ‘‘Cap’’ on the tip of Salmonella flagella. J. Mol.

Biol. 184, 735–737.Ikeda, T., Homma, M., Iino, T., Asakura, S., and Kamiya, R. (1987). Localization and

stoichiometry of hook-associated proteins within Salmonella typhimurium flagella. J.Bacteriol. 169, 1168–1173.

Ikeda, T., Yamaguchi, S., and Hotani, H. (1993). Flagellar growth in a filament-lessSalmonella fliD mutant supplemented with purified hook-associated protein-2. J. Biochem.(Tokyo) 114, 39–44.

Imada, K., Minamino, T., Tahara, A., and Namba, K. (2007). Structural similarity betweenthe flagellar type III ATPase FliI and F1-ATPase subunits. Proc. Natl. Acad. Sci. USA 104,485–490.

Imae, Y., and Atsumi, T. (1989). Naþ-driven bacterial flagellar motors. J. Bioenerg. Bio-membr. 21, 705–716.

Inoue, Y., Lo, C. J., Fukuoka, H., Takahashi, H., Sowa, Y., Pilizota, T., Wadhams, G. H.,Homma, M., Berry, R. M., and Ishijima, A. (2008). Torque-speed relationships of Naþ-driven chimeric flagellar motors in Escherichia coli. J. Mol. Biol. 376, 1251–1259.

Ito, M., Hicks, D. B., Henkin, T. M., Guffanti, A. A., Powers, B. D., Zvi, L., Uematsu, K.,and Krulwich, T. A. (2004). MotPS is the stator-force generator for motility of alkali-philic Bacillus, and its homologue is a second functional Mot in Bacillus subtilis. Mol.Microbiol. 53, 1035–1049.

Ito, M., Terahara, N., Fujinami, S., and Krulwich, T. A. (2005). Properties of motility inBacillus subtilis powered by the Hþ-coupled MotAB flagellar stator, Naþ-coupled MotPSor hybrid stators MotAS or MotPB. J. Mol. Biol. 352, 396–408.

Iwazawa, J., Imae, Y., and Kobayasi, S. (1993). Study of the torque of the bacterial flagellarmotor using a rotating electric field. Biophys. J. 64, 925–933.

Jaques, S., Kim, Y. K., and McCarter, L. L. (1999). Mutations conferring resistance tophenamil and amiloride, inhibitors of sodium-driven motility of Vibrio parahaemolyticus.Proc. Natl. Acad. Sci. USA 96, 5740–5745.

Kamiya, R., and Asakura, S. (1976). Helical transformations of Salmonella flagella in vitro.J. Mol. Biol. 106, 167–186.

Karlinsey, J. E., Pease, A. J., Winkler, M. E., Bailey, J. L., and Hughes, K. T. (1997). The flkgene of Salmonella typhimurium couples flagellar P- and L-ring assembly to flagellarmorphogenesis. J. Bacteriol. 179, 2389–2400.

Katayama, E., Shiraishi, T., Oosawa, K., Baba, N., and Aizawa, S. (1996). Geometry of theflagellar motor in the cytoplasmic membrane of Salmonella typhimurium as determined bystereo-photogrammetry of quick-freeze deep-etch replica images. J. Mol. Biol. 255,458–475.

Khan, S., and Berg, H. C. (1983). Isotope and thermal effects in chemiosmotic coupling tothe flagellar motor of Streptococcus. Cell 32, 913–919.

Khan, S., Dapice, M., and Reese, T. S. (1988). Effects of mot gene expression on thestructure of the flagellar motor. J. Mol. Biol. 202, 575–584.

Kobayashi, K., Saitoh, T., Shah, D. S., Ohnishi, K., Goodfellow, I. G., Sockett, R. E., andAizawa, S. I. (2003). Purification and characterization of the flagellar basal body ofRhodobacter sphaeroides. J. Bacteriol. 185, 5295–5300.

Koebnik, R. (1995). Proposal for a peptidoglycan-associating alpha-helical motif in theC-terminal regions of some bacterial cell-surface proteins.Mol. Microbiol. 16, 1269–1270.

Kojima, S., and Blair, D. F. (2001). Conformational change in the stator of the bacterialflagellar motor. Biochemistry 40, 13041–13050.


Kojima, S., and Blair, D. F. (2004a). The bacterial flagellar motor: Structure and function of acomplex molecular machine. Int. Rev. Cytol. 233, 93–134.

Kojima, S., and Blair, D. F. (2004b). Solubilization and purification of the MotA/MotBcomplex of Escherichia coli. Biochemistry 43, 26–34.

Kojima, S., Atsumi, T., Muramoto, K., Kudo, S., Kawagishi, I., and Homma, M. (1997).Vibrio alginolyticusmutants resistant to phenamil, a specific inhibitor of the sodium-drivenflagellar motor. J. Mol. Biol. 265, 310–318.

Kojima, S., Asai, Y., Atsumi, T., Kawagishi, I., and Homma, M. (1999). Naþ-drivenflagellar motor resistant to phenamil, an amiloride analog, caused by mutations in putativechannel components. J. Mol. Biol. 285, 1537–1547.

Kojima, S., Shoji, T., Asai, Y., Kawagishi, I., and Homma, M. (2000). A slow-motilityphenotype caused by substitutions at residue Asp31 in the PomA channel component of asodium-driven flagellar motor. J. Bacteriol. 182, 3314–3318.

Kojima, M., Kubo, R., Yakushi, T., Homma, M., and Kawagishi, I. (2007). The bidirec-tional polar and unidirectional lateral flagellar motors of Vibrio alginolyticus are controlledby a single CheY species. Mol. Microbiol. 64, 57–67.

Kojima, S., Shinohara, A., Terashima, H., Yakushi, T., Sakuma, M., Homma, M.,Namba, K., and Imada, K. (2008). Insights into the stator assembly of the Vibrio flagellarmotor from the crystal structure of MotY. Proc. Natl. Acad. Sci. USA 190, 7696–7701.

Kubori, T., Shimamoto, N., Yamaguchi, S., Namba, K., and Aizawa, S. (1992). Morpho-logical pathway of flagellar assembly in Salmonella typhimurium. J. Mol. Biol. 226,433–446.

Kudo, S., Magariyama, Y., and Aizawa, S.-I. (1990). Abrupt changes in flagellar rotationobserved by laser dark-field microscopy. Nature 346, 677–680.

Kusumoto, A., Kamisaka, K., Yakushi, T., Terashima, H., Shinohara, A., and Homma, M.(2006). Regulation of polar flagellar number by the flhF and flhG genes in Vibrioalginolyticus. J. Biochem. (Tokyo) 139, 113–121.

Kusumoto, A., Shinohara, A., Terashima, H., Kojima, S., Yakushi, T., and Homma, M.(2008). Collaboration of FlhF and FlhG to regulate polar-flagella number and localizationin Vibrio alginolyticus. Microbiology 154, 1390–1399.

Kutsukake, K. (1994). Excretion of the anti-sigma factor through a flagellar substructurecouples flagellar gene expression with flagellar assembly in Salmonella typhimurium. Mol.Gen. Genet. 243, 605–612.

Kutsukake, K. (1997). Hook-length control of the export-switching machinery involves adouble-locked gate in Salmonella typhimurium flagellar morphogenesis. J. Bacteriol. 179,1268–1273.

Kutsukake, K., and Iino, T. (1994). Role of the FliA-FlgM regulatory system on thetranscriptional control of the flagellar regulon and flagellar formation in Salmonellatyphimurium. J. Bacteriol. 176, 3598–3605.

Kutsukake, K., Ohya, Y., and Iino, T. (1990). Transcriptional analysis of the flagellar regulonof Salmonella typhimurium. J. Bacteriol. 172, 741–747.

Kutsukake, K., Minamino, T., and Yokoseki, T. (1994). Isolation and characterization ofFliK-independent flagellation mutants from Salmonella typhimurium. J. Bacteriol. 176,7625–7629.

Leake, M. C., Chandler, J. H., Wadhams, G. H., Bai, F., Berry, R. M., and Armitage, J. P.(2006). Stoichiometry and turnover in single, functioning membrane protein complexes.Nature 443, 355–358.

Lee, S. Y., Cho, H. S., Pelton, J. G., Yan, D., Henderson, R. K., King, D. S., Huang, L.,Kustu, S., Berry, E. A., and Wemmer, D. E. (2001). Crystal structure of an activatedresponse regulator bound to its target. Nat. Struct. Biol. 8, 52–56.

Liu, X., and Matsumura, P. (1994). The FlhD/FlhC complex, a transcriptional activator ofthe Escherichia coli flagellar class II operons. J. Bacteriol. 176, 7345–7351.


Lloyd, S. A., and Blair, D. F. (1997). Charged residues of the rotor protein FliGessential for torque generation in the flagellar motor of Escherichia coli. J. Mol. Biol. 266,733–744.

Lloyd, S. A., Tang, H., Wang, X., Billings, S., and Blair, D. F. (1996). Torque generation inthe flagellar motor of Escherichia coli: Evidence of a direct role for FliG but not for FliM orFliN. J. Bacteriol. 178, 223–231.

Lloyd, S. A., Whitby, F. G., Blair, D. F., and Hill, C. P. (1999). Structure of the C-terminaldomain of FliG, a component of the rotor in the bacterial flagellar motor. Nature 400,472–475.

Lo, C. J., Leake, M. C., and Berry, R. M. (2006). Fluorescence measurement of intracellularsodium concentration in single Escherichia coli cells. Biophys. J. 90, 357–365.

Lo, C. J., Leake, M. C., Pilizota, T., and Berry, R. M. (2007). Nonequivalence ofmembrane voltage and ion-gradient as driving forces for the bacterial flagellar motor atlow load. Biophys. J. 93, 294–302.

Lowe, G., Meister, M., and Berg, H. C. (1987). Rapid rotation of flagellar bundles inswimming bacteria. Nature 325, 637–640.

Luirink, J., and Sinning, I. (2004). SRP-mediated protein targeting: Structure and functionrevisited. Biochim. Biophys. Acta 1694, 17–35.

Macnab, R. M. (2003). How bacteria assemble flagella. Annu. Rev. Microbiol. 57, 77–100.Macnab, R. M. (2004). Type III flagellar protein export and flagellar assembly. Biochim.

Biophys. Acta 1694, 207–217.Macnab, R. M., and Ornston, M. K. (1977). Normal-to-curly flagellar transitions and their

role in bacterial tumbling. Stabilization of an alternative quaternary structure by mechan-ical force. J. Mol. Biol. 112, 1–30.

Manson, M. D., Tedesco, P. M., and Berg, H. C. (1980). Energetics of flagellar rotation inbacteria. J. Mol. Biol. 138, 541–561.

Marykwas, D. L., Schmidt, S. A., and Berg, H. C. (1996). Interacting components of theflagellar motor of Escherichia coli revealed by the two-hybrid system in yeast. J. Mol. Biol.256, 564–576.

Mathews, M. A., Tang, H. L., and Blair, D. F. (1998). Domain analysis of the FliM proteinof Escherichia coli. J. Bacteriol. 180, 5580–5590.

McCarter, L. L. (1994a). MotX, the channel component of the sodium-type flagellar motor.J. Bacteriol. 176, 5988–5998.

McCarter, L. L. (1994b). MotY, a component of the sodium-type flagellar motor. J. Bacteriol.176, 4219–4225.

McCarter, L. L. (2001). Polar flagellar motility of the Vibrionaceae.Microbiol. Mol. Biol. Rev.65, 445–462.

McCarter, L. L. (2004). Dual flagellar systems enable motility under different circumstances.J. Mol. Microbiol. Biotechnol. 7, 18–29.

McCarter, L., Hilmen, M., and Silverman, M. (1988). Flagellar dynamometer controlswarmer cell differentiation of V. parahaemolyticus. Cell 54, 345–351.

McMurry, J. L., Murphy, J. W., and Gonzalez-Pedrajo, B. (2006). The FliN-FliH interac-tion mediates localization of flagellar export ATPase FliI to the C ring complex. Biochem-istry 45, 11790–11798.

Mehta, A. D., Rock, R. S., Rief, M., Spudich, J. A., Mooseker, M. S., and Cheney, R. E.(1999). Myosin-V is a processive actin-based motor. Nature 400, 590–593.

Mimori, Y., Yamashita, I., Murata, K., Fujiyoshi, Y., Yonekura, K., Toyoshima, C., andNamba, K. (1995). The structure of the R-type straight flagellar filament of Salmonella at9 angstrom resolution by electron cryomicroscopy. J. Mol. Biol. 249, 69–87.

Minamino, T., and Macnab, R. M. (1999). Components of the Salmonellaflagellar export apparatus and classification of export substrates. J. Bacteriol. 181,1388–1394.


Minamino, T., and Macnab, R. M. (2000a). FliH, a soluble component of the type IIIflagellar export apparatus of Salmonella, forms a complex with FliI and inhibits its ATPaseactivity. Mol. Microbiol. 37, 1494–1503.

Minamino, T., and Macnab, R. M. (2000b). Interactions among components of theSalmonella flagellar export apparatus and its substrates. Mol. Microbiol. 35, 1052–1064.

Minamino, T., and Namba, K. (2004). Self-assembly and type III protein export of thebacterial flagellum. J. Mol. Microbiol. Biotechnol. 7, 5–17.

Minamino, T., Doi, H., and Kutsukake, K. (1999). Substrate specificity switching of theflagellum-specific export apparatus during flagellar morphogenesis in Salmonella typhimur-ium. Biosci. Biotechnol. Biochem. 63, 1301–1303.

Minamino, T., Chu, R., Yamaguchi, S., andMacnab, R. M. (2000a). Role of FliJ in flagellarprotein export in Salmonella. J. Bacteriol. 182, 4207–4215.

Minamino, T., Yamaguchi, S., and Macnab, R. M. (2000b). Interaction between FliE andFlgB, a proximal rod component of the flagellar basal body of Salmonella. J. Bacteriol. 182,3029–3036.

Minamino, T., Kazetani, K., Tahara, A., Suzuki, H., Furukawa, Y., Kihara, M., andNamba, K. (2006). Oligomerization of the bacterial flagellar ATPase FliI is controlledby its extreme N-terminal region. J. Mol. Biol. 360, 510–519.

Morgan, D. G., Owen, C., Melanson, L. A., and Derosier, D. J. (1995). Structure ofbacterial flagellar filaments at 11 angstrom resolution: Packing of the alpha-helices.J. Mol. Biol. 249, 88–110.

Moriya, N., Minamino, T., Hughes, K. T., Macnab, R. M., and Namba, K. (2006). Thetype III flagellar export specificity switch is dependent on FliK ruler and a molecularclock. J. Mol. Biol. 359, 466–477.

Murata, T., Yamato, I., Kakinuma, Y., Leslie, A. G., and Walker, J. E. (2005). Structure ofthe rotor of the V-Type Naþ-ATPase from Enterococcus hirae. Science 308, 654–659.

Murphy, G. E., Leadbetter, J. R., and Jensen, G. J. (2006). In situ structure of the completeTreponema primitia flagellar motor. Nature 442, 1062–1064.

Nambu, T., and Kutsukake, K. (2000). The Salmonella FlgA protein, a putativeve periplas-mic chaperone essential for flagellar P ring formation. Microbiology 146, 1171–1178.

Nambu, T., Minamino, T., Macnab, R. M., and Kutsukake, K. (1999). Peptidoglycan-hydrolyzing activity of the FlgJ protein, essential for flagellar rod formation in Salmonellatyphimurium. J. Bacteriol. 181, 1555–1561.

Obara, M., Yakushi, T., Kojima, S., and Homma, M. (2008). Roles of charged residues inthe C-terminal region of PomA, a stator component of the Na+-driven flagellar motor.J. Bacteriol. 190, 3565–3571.

Ohnishi, K., Kusukake, K., Suzuki, H., and Iino, T. (1990). Gene FliA encodes an alterna-tive sigma factor specific for flagellar operons in Salmonella typhimurium.Mol. Gen. Genet.221, 139–147.

Ohnishi, K., Kutsukake, K., Suzuki, H., and Iino, T. (1992). A novel transcriptionalregulation mechanism in the flagellar regulon of Salmonella typhimurium—An anti-sigma factor inhibits the activity of the flagellum-specific sigma factor, sigmaF. Mol.Microbiol. 6, 3149–3157.

Okabe, M., Yakushi, T., Asai, Y., and Homma, M. (2001). Cloning and characterization ofmotX, a Vibrio alginolyticus sodium-driven flagellar motor gene. J. Biochem. (Tokyo) 130,879–884.

Okabe, M., Yakushi, T., and Homma, M. (2005). Interactions of MotX with MotY andwith the PomA/PomB sodium ion channel complex of the Vibrio alginolyticus polarflagellum. J. Biol. Chem. 280, 25659–25664.

Okunishi, I., Kawagishi, I., and Homma, M. (1996). Cloning and characterization of motY,a gene coding for a component of the sodium-driven flagellar motor inVibrio alginolyticus.J. Bacteriol. 178, 2409–2415.


Oosawa, K., Ueno, T., and Aizawa, S. (1994). Overproduction of the bacterial flagellarswitch proteins and their interactions with the MS ring complex in vitro. J. Bacteriol. 176,3683–3691.

Pandza, S., Baetens, M., Park, C. H., Au, T., Keyhan, M., and Matin, A. (2000). TheG-protein FlhF has a role in polar flagellar placement and general stress responseinduction in Pseudomonas putida. Mol. Microbiol. 36, 414–423.

Park, S. Y., Lowder, B., Bilwes, A. M., Blair, D. F., and Crane, B. R. (2006). Structure ofFliM provides insight into assembly of the switch complex in the bacterial flagella motor.Proc. Natl. Acad. Sci. USA 103, 11886–11891.

Parkinson, J. S. (2003). Bacterial chemotaxis: A new player in response regulator dephos-phorylation. J. Bacteriol. 185, 1492–1494.

Parkinson, J. S., Ames, P., and Studdert, C. A. (2005). Collaborative signaling by bacterialchemoreceptors. Curr. Opin. Microbiol. 8, 116–121.

Paul, K., and Blair, D. F. (2006). Organization of FliN subunits in the flagellar motor ofEscherichia coli. J. Bacteriol. 188, 2502–2511.

Paul, K., Harmon, J. G., and Blair, D. F. (2006). Mutational analysis of the flagellar rotorprotein FliN: Identification of surfaces important for flagellar assembly and switching.J. Bacteriol. 188, 5240–5248.

Platzer, J., Sterr, W., Hausmann, M., and Schmitt, R. (1997). Three genes of a motilityoperon and their role in flagellar rotary speed variation in Rhizobium meliloti. J. Bacteriol.179, 6391–6399.

Pruss, B. M., and Matsumura, P. (1996). A regulator of the flagellar regulon of Escherichia coli,flhD, also affects cell division. J. Bacteriol. 178, 668–674.

Pruss, B. M., and Matsumura, P. (1997). Cell cycle regulation of flagellar genes. J. Bacteriol.179, 5602–5604.

Reid, S. W., Leake, M. C., Chandler, J. H., Lo, C. J., Armitage, J. P., and Berry, R. M.(2006). The maximum number of torque-generating units in the flagellar motor ofEscherichia coli is at least 11. Proc. Natl. Acad. Sci. USA 103, 8066–8071.

Ryu, W. S., Berry, R. M., and Berg, H. C. (2000). Torque-generating units of the flagellarmotor of Escherichia coli have a high duty ratio. Nature 403, 444–447.

Sagi, Y., Khan, S., and Eisenbach, M. (2003). Binding of the chemotaxis response regulatorCheY to the isolated, intact switch complex of the bacterial flagellar motor: Lack ofcooperativity. J. Biol. Chem. 278, 25867–25871.

Samatey, F. A., Imada, K., Nagashima, S., Vonderviszt, F., Kumasaka, T., Yamamoto, M.,and Namba, K. (2001). Structure of the bacterial flagellar protofilament and implicationsfor a switch for supercoiling. Nature 410, 331–337.

Samatey, F. A., Matsunami, H., Imada, K., Nagashima, S., Shaikh, T. R., Thomas, D. R.,Chen, J. Z., Derosier, D. J., Kitao, A., and Namba, K. (2004). Structure of the bacterialflagellar hook and implication for the molecular universal joint mechanism. Nature 431,1062–1068.

Samuel, A. D., and Berg, H. C. (1995). Fluctuation analysis of rotational speeds of thebacterial flagellar motor. Proc. Natl. Acad. Sci. USA 92, 3502–3506.

Samuel, A. D., and Berg, H. C. (1996). Torque-generating units of the bacterial flagellarmotor step independently. Biophys. J. 71, 918–923.

Sato, K., and Homma, M. (2000a). Functional reconstitution of the Naþ-driven polarflagellar motor component of Vibrio alginolyticus. J. Biol. Chem. 275, 5718–5722.

Sato, K., and Homma, M. (2000b). Multimeric structure of PomA, a component ofthe Naþ-driven polar flagellar motor of Vibrio alginolyticus. J. Biol. Chem. 275,20223–20228.

Schmitt, R. (2002). Sinorhizobial chemotaxis: A departure from the enterobacterial para-digm. Microbiology 148, 627–631.


Schmitt, R. (2003). Helix rotation model of the flagellar rotary motor. Biophys. J. 85,843–852.

Schnitzer, M. J., and Block, S. M. (1997). Kinesin hydrolyses one ATP per 8-nm step.Nature388, 386–390.

Schoenhals, G. J., and Macnab, R. M. (1996). Physiological and biochemical analyses ofFlgH, a lipoprotein forming the outer membrane L ring of the flagellar basal body ofSalmonella typhimurium. J. Bacteriol. 178, 4200–4207.

Shah, D. S., and Sockett, R. E. (1995). Analysis of the motA flagellar motor gene fromRhodobacter sphaeroides, a bacterium with a unidirectional, stop-start flagellum. Mol.Microbiol. 17, 961–969.

Shapiro, L., McAdams, H. H., and Losick, R. (2002). Generating and exploiting polarity inbacteria. Science 298, 1942–1946.

Shibata, S., Takahashi, N., Chevance, F. F., Karlinsey, J. E., Hughes, K. T., and Aizawa, S.(2007). FliK regulates flagellar hook length as an internal ruler. Mol. Microbiol. 64,1404–1415.

Silverman, M., and Simon, M. I. (1974). Flagellar rotation and the mechanism of bacterialmotility. Nature 249, 73–74.

Sockett, H., Yamaguchi, S., Kihara, M., Irikura, V., and Macnab, R. M. (1992). Molecularanalysis of the flagellar switch protein FliM of Salmonella typhimurium. J. Bacteriol. 174,793–806.

Sourjik, V., and Berg, H. C. (2002). Binding of the Escherichia coli response regulator CheYto its target measured in vivo by fluorescence resonance energy transfer. Proc. Natl. Acad.Sci. USA 99, 12669–12674.

Sowa, Y., Hotta, H., Homma, M., and Ishijima, A. (2003). Torque-speed relationship of theNaþ-driven flagellar motor of Vibrio alginolyticus. J. Mol. Biol. 327, 1043–1051.

Sowa, Y., Rowe, A. D., Leake, M. C., Yakushi, T., Homma, M., Ishijima, A., andBerry, R. M. (2005). Direct observation of steps in rotation of the bacterial flagellarmotor. Nature 437, 916–919.

Stader, J., Matsumura, P., Vacante, D., Dean, G. E., and Macnab, R. M. (1986). Nucleotidesequence of the Escherichia coli MotB gene and site-limited incorporation of its productinto the cytoplasmic membrane. J. Bacteriol. 166, 244–252.

Stallmeyer, B. M. J., Aizawa, S., Macnab, R. M., and DeRosier, D. J. (1989). Imagereconstruction of the flagellar basal body of Salmonella typhimurium. J. Mol. Biol. 205,519–528.

Stewart, B. J., and McCarter, L. L. (2003). Lateral flagellar gene system of Vibrio parahaemo-lyticus. J. Bacteriol. 185, 4508–4518.

Stolz, B., and Berg, H. C. (1991). Evidence for interactions between MotA and MotB,torque-generating elements of the flagellar motor of Escherichia coli. J. Bacteriol. 173,7033–7037.

Suzuki, H., Yonekura, K., and Namba, K. (2004). Structure of the rotor of the bacterialflagellar motor revealed by electron cryomicroscopy and single-particle image analysis.J. Mol. Biol. 337, 105–113.

Suzuki, T., and Komeda, Y. (1981). Incomplete flagellar structures in Escherichia colimutants.J. Bacteriol. 145, 1036–1041.

Suzuki, T., Iino, T., Horiguchi, T., and Yamaguchi, S. (1978). Incomplete flagellar struc-tures in nonflagellate mutants of Salmonella typhimurium. J. Bacteriol. 133, 904–915.

Tang, H., Billings, S., Wang, X., Sharp, L., and Blair, D. F. (1995). Regulated under-expression and overexpression of the FliN protein of Escherichia coli and evidence for aninteraction between FliN and FliM in the flagellar motor. J. Bacteriol. 177, 3496–3503.

Tang, H., Braun, T. F., and Blair, D. F. (1996). Motility protein complexes in the bacterialflagellar motor. J. Mol. Biol. 261, 209–221.


Terashima, H., Fukuoka, H., Yakushi, T., Kojima, S., and Homma, M. (2006). The Vibriomotor proteins, MotX and MotY, are associated with the basal body of Na+-drivenflagella and required for stator formation. Mol. Microbiol. 62, 1170–1180.

Thomas, D. R., Morgan, D. G., and DeRosier, D. J. (1999). Rotational symmetry of the Cring and a mechanism for the flagellar rotary motor. Proc. Natl. Acad. Sci. USA 96,10134–10139.

Thomas, D. R., Morgan, D. G., and DeRosier, D. J. (2001). Structures of bacterial flagellarmotors from two FliF-FliG gene fusion mutants. J. Bacteriol. 183, 6404–6412.

Thomas, D. R., Francis, N. R., Xu, C., and DeRosier, D. J. (2006). The three-dimensionalstructure of the flagellar rotor from a clockwise-locked mutant of Salmonella entericaserovar Typhimurium. J. Bacteriol. 188, 7039–7048.

Toker, A. S., and Macnab, R. M. (1997). Distinct regions of bacterial flagellar switch proteinFliM interact with FliG, FliN and CheY. J. Mol. Biol. 273, 623–634.

Toutain, C. M., Zegans, M. E., and O’Toole, G. A. (2005). Evidence for two flagellar statorsand their role in the motility of Pseudomonas aeruginosa. J. Bacteriol. 187, 771–777.

Ueno, T., Oosawa, K., and Aizawa, S. (1992). M ring, S ring and proximal rod of theflagellar basal body of Salmonella typhimurium are composed of subunits of a single protein,FliF. J. Mol. Biol. 227, 672–677.

Vogler, A. P., Homma, M., Irikura, V. M., and Macnab, R. M. (1991). Salmonella typhimur-ium mutants defective in flagellar filament regrowth and sequence similarity of FliI toFOF1, vacuolar, and archaebacterial ATPase subunits. J. Bacteriol. 173, 3564–3572.

Waters, R. C., O’Toole, P. W., and Ryan, K. A. (2007). The FliK protein and flagellarhook-length control. Protein Sci. 16, 769–780.

Welch, M., Oosawa, K., Aizawa, S. I., and Eisenbach, M. (1993). Phosphorylation-depen-dent binding of a signal molecule to the flagellar switch of bacteria. Proc. Natl. Acad. Sci.USA 90, 8787–8791.

Wilson, M. L., and Macnab, R. M. (1988). Overproduction of the MotA protein ofEscherichia coli and estimation of its wild-type level. J. Bacteriol. 170, 588–597.

Wilson, M. L., and Macnab, R. M. (1990). Co-overproduction and localization of theEscherichia coli motility proteins MotA and MotB. J. Bacteriol. 172, 3932–3939.

Wolfe, A. J., and Visick, K. L. (2008). Get the message out: Cyclic-di-GMP regulatesmultiple levels of flagellar-based motility. J. Bacteriol. 190, 463–475.

Wu, J., and Newton, A. (1997). Regulation of the Caulobacter flagellar gene hierarchy; notjust for motility. Mol. Microbiol. 24, 233–239.

Xing, J., Bai, F., Berry, R., and Oster, G. (2006). Torque-speed relationship of the bacterialflagellar motor. Proc. Natl. Acad. Sci. USA 103, 1260–1265.

Yakushi, T., Maki, S., and Homma, M. (2004). Interaction of PomB with the thirdtransmembrane segment of PomA in the Naþ-driven polar flagellum of Vibrio alginoly-ticus. J. Bacteriol. 186, 5281–5291.

Yakushi, T., Yang, J., Fukuoka, H., Homma, M., and Blair, D. F. (2006). Roles of chargedresidues of rotor and stator in flagellar rotation: Comparative study using Hþ-driven andNaþ-driven motors in Escherichia coli. J. Bacteriol. 188, 1466–1472.

Yamaguchi, S., Aizawa, S., Kihara, M., Isomura, M., Jones, C. J., and Macnab, R. M.(1986a). Genetic evidence for a switching and energy-transducing complex in theflagellar motor of Salmonella typhimurium. J. Bacteriol. 168, 1172–1179.

Yamaguchi, S., Fujita, H., Ishihara, H., Aizawa, S., andMacnab, R. M. (1986b). Subdivisionof flagellar genes of Salmonella typhimurium into regions responsible for assembly, rotation,and switching. J. Bacteriol. 166, 187–193.

Yamashita, A., Singh, S. K., Kawate, T., Jin, Y., and Gouaux, E. (2005). Crystal structure ofa bacterial homologue of Naþ/Cl�-dependent neurotransmitter transporters. Nature437, 215–223.


Yamashita, I., Hasegawa, K., Suzuki, H., Vonderviszt, F., Mimori-Kiyosue, Y., andNamba, K. (1998). Structure and switching of bacterial flagellar filaments studied byX-ray fiber diffraction. Nat. Struct. Biol. 5, 125–132.

Yasuda, R., Noji, H., Kinosita, K., Jr., and Yoshida, M. (1998). F1-ATPase is a highlyefficient molecular motor that rotates with discrete 120 degree steps.Cell 93, 1117–1124.

Yonekura, K., Maki, S., Morgan, D. G., DeRosier, D. J., Vonderviszt, F., Imada, K., andNamba, K. (2000). The bacterial flagellar cap as the rotary promoter of flagellin self-assembly. Science 290, 2148–2152.

Yonekura, K., Maki-Yonekura, S., and Namba, K. (2003). Complete atomic model of thebacterial flagellar filament by electron cryomicroscopy. Nature 424, 643–650.

Yonekura, K., Yakushi, T., Atsumi, T., Maki-Yonekura, S., Homma, M., and Namba, K.(2006). Electron cryomicroscopic visualization of PomA/B stator units of the sodium-driven flagellar motor in liposomes. J. Mol. Biol. 357, 73–81.

Yorimitsu, T., and Homma, M. (2001). Naþ-driven flagellar motor of Vibrio. Biochim.Biophys. Acta 1505, 82–93.

Yorimitsu, T., Asai, Y., Sato, K., and Homma, M. (2000). Intermolecular cross-linkingbetween the periplasmic Loop3–4 regions of PomA, a component of the Naþ-drivenflagellar motor of Vibrio alginolyticus. J. Biol. Chem. 275, 31387–31391.

Yorimitsu, T., Sowa, Y., Ishijima, A., Yakushi, T., and Homma, M. (2002). The systematicsubstitutions around the conserved charged residues of the cytoplasmic loop of Naþ-driven flagellar motor component PomA. J. Mol. Biol. 320, 403–413.

Yorimitsu, T., Mimaki, A., Yakushi, T., and Homma, M. (2003). The conserved chargedresidues of the C-terminal region of FliG, a rotor component of the Naþ-driven flagellarmotor. J. Mol. Biol. 334, 567–583.

Yorimitsu, T., Kojima, M., Yakushi, T., and Homma, M. (2004). Multimeric structure ofthe PomA/PomB channel complex in the Naþ-driven flagellar motor of Vibrio alginoly-ticus. J. Biochem. (Tokyo) 135, 43–51.

Zhao, R., Amsler, C. D., Matsumura, P., and Khan, S. (1996a). FliG and FliM distributionin the Salmonella typhimurium cell and flagellar basal bodies. J. Bacteriol. 178, 258–265.

Zhao, R., Pathak, N., Jaffe, H., Reese, T. S., and Khan, S. (1996b). FliN is a major structuralprotein of the C-ring in the Salmonella typhimurium flagellar basal body. J. Mol. Biol. 261,195–208.

Zhou, J., and Blair, D. F. (1997). Residues of the cytoplasmic domain of MotA essential fortorque generation in the bacterial flagellar motor. J. Mol. Biol. 273, 428–439.

Zhou, J., Fazzio, R. T., and Blair, D. F. (1995). Membrane topology of the MotA protein ofEscherichia coli. J. Mol. Biol. 251, 237–242.

Zhou, J., Lloyd, S. A., and Blair, D. F. (1998a). Electrostatic interactions between rotor andstator in the bacterial flagellar motor. Proc. Natl. Acad. Sci. USA 95, 6436–6441.

Zhou, J., Sharp, L. L., Tang, H. L., Lloyd, S. A., Billings, S., Braun, T. F., and Blair, D. F.(1998b). Function of protonatable residues in the flagellar motor of Escherichia coli: Acritical role for Asp 32 of MotB. J. Bacteriol. 180, 2729–2735.

Chapter 2 - Flagellar Motility in Bacteria: Structure and...

Documents

Transcript of Chapter 2 - Flagellar Motility in Bacteria: Structure and...